paper_id
uint32
title
string
authors
list
ecva_url
string
pdf_url
string
supp_url
string
doi
string
arxiv_id
string
arxiv_id_source
string
abstract
large_string
embedding
list
0
Quaternion Equivariant Capsule Networks for 3D Point Clouds
[ "Yongheng Zhao", "Tolga Birdal", "Jan Eric Lenssen", "Emanuele Menegatti", "Leonidas Guibas", "Federico Tombari" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/267_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460001.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460001-supp.pdf
10.1007/978-3-030-58452-8_1
1912.12098
title_snapshot
We present a 3D capsule module for processing point clouds that is equivariant to 3D rotations and translations, as well as invariant to permutations of the input points. The operator receives a sparse set of local reference frames, computed from an input point cloud and establishes end-to-end transformation equivarian...
[ 0.027841586619615555, -0.005717919208109379, 0.0068319630809128284, 0.026674972847104073, 0.010356323793530464, 0.031612880527973175, -0.00901066418737173, 0.0266374833881855, -0.0387546680867672, -0.052526332437992096, -0.035126712173223495, -0.04663495346903801, -0.05695498734712601, 0.0...
1
DeepFit: 3D Surface Fitting via Neural Network Weighted Least Squares
[ "Yizhak Ben-Shabat", "Stephen Gould" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/283_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460018.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460018-supp.zip
10.1007/978-3-030-58452-8_2
2003.10826
title_snapshot
We propose a surface fitting method for unstructured 3D point clouds. This method, called DeepFit, incorporates a neural network to learn point-wise weights for weighted least squares polynomial surface fitting. The learned weights act as a soft selection for the neighborhood of surface points thus avoiding the scale s...
[ 0.017680702731013298, -0.011651786044239998, 0.02058914490044117, 0.01640520989894867, 0.03924676403403282, 0.08270272612571716, -0.000447286874987185, -0.017930785194039345, -0.008494812995195389, -0.07081814855337143, -0.03445448353886604, 0.016805635765194893, -0.0650273859500885, 0.008...
2
NSGANetV2: Evolutionary Multi-Objective Surrogate-Assisted Neural Architecture Search
[ "Zhichao Lu", "Kalyanmoy Deb", "Erik Goodman", "Wolfgang Banzhaf", "Vishnu Naresh Boddeti" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/343_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460035.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460035-supp.pdf
10.1007/978-3-030-58452-8_3
2007.10396
title_snapshot
In this paper, we propose an efficient NAS algorithm for generating task-specific models that are competitive under multiple competing objectives. It comprises of two surrogates, one at the architecture level to improve sample efficiency and one at the weights level, through a supernet, to improve gradient descent trai...
[ -0.013892171904444695, -0.03893627971410751, -0.01785079762339592, 0.0331866554915905, 0.02768395096063614, 0.059083547443151474, 0.040521614253520966, -0.011409825645387173, -0.026806404814124107, -0.05608309060335159, 0.00863640010356903, -0.016208436340093613, -0.05706172436475754, -0.0...
3
Describing Textures using Natural Language
[ "Chenyun Wu", "Mikayla Timm", "Subhransu Maji" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/384_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460052.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460052-supp.pdf
10.1007/978-3-030-58452-8_4
2008.01180
title_snapshot
Textures in natural images can be characterized by color, shape, periodicity of elements within them, and other attributes that can be described using natural language. In this paper, we study the problem of describing visual attributes of texture on a novel dataset containing rich descriptions of textures, and conduct...
[ 0.005807932466268539, -0.0025562152732163668, -0.009576418437063694, 0.057597629725933075, 0.031093228608369827, 0.03307843580842018, 0.012071436271071434, 0.03428317606449127, -0.015688760206103325, -0.04486888274550438, -0.07443780452013016, 0.00843032170087099, -0.05479327589273453, 0.0...
4
Empowering Relational Network by Self-Attention Augmented Conditional Random Fields for Group Activity Recognition
[ "Rizard Renanda Adhi Pramono", "Yie Tarng Chen", "Wen Hsien Fang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/410_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460069.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460069-supp.pdf
10.1007/978-3-030-58452-8_5
null
null
This paper presents a novel relational network for group activity recognition. The core of our network is to augment the conditional random fields (CRF), amenable to learning inter-dependency of correlated observations, with the newly devised temporal and spatial self-attention to learn the temporal evolution and spati...
[ 0.016397925093770027, -0.004613704048097134, 0.00811392068862915, 0.03237757831811905, 0.01852143183350563, 0.009811454452574253, 0.030547864735126495, 0.020478639751672745, -0.02628106251358986, -0.01805904693901539, -0.018144547939300537, -0.00030932301888242364, -0.05106184259057045, -0...
5
AiR: Attention with Reasoning Capability
[ "Shi Chen", "Ming Jiang", "Jinhui Yang", "Qi Zhao" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/445_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460086.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460086-supp.zip
10.1007/978-3-030-58452-8_6
2007.14419
title_snapshot
While attention has been an increasingly popular component in deep neural networks to both interpret and boost performance of models, little work has examined how attention progresses to accomplish a task and whether it is reasonable. In this work, we propose an Attention with Reasoning capability (AiR) framework that ...
[ 0.015393841080367565, 0.008875361643731594, 0.00937971007078886, 0.021606547757983208, 0.0459681935608387, 0.01329716108739376, 0.03824016451835632, 0.025716377422213554, -0.024477452039718628, -0.0134524405002594, -0.03726742044091225, 0.025565385818481445, -0.06493142247200012, -0.019508...
6
Self6D: Self-Supervised Monocular 6D Object Pose Estimation
[ "Gu Wang", "Fabian Manhardt", "Jianzhun Shao", "Xiangyang Ji", "Nassir Navab", "Federico Tombari" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/500_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460103.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460103-supp.zip
10.1007/978-3-030-58452-8_7
2004.06468
title_snapshot
6D object pose estimation is a fundamental problem in computer vision. Convolutional Neural Networks (CNNs) have recently proven to be capable of predicting reliable 6D pose estimates even from monocular images. Nonetheless, CNNs are identified as being extremely data-driven, and acquiring adequate annotations is often...
[ 0.0031111540738493204, -0.004617484752088785, -0.019580397754907608, 0.04967383295297623, 0.014394273050129414, 0.05891015753149986, 0.005987190641462803, 0.005565831903368235, -0.03740601986646652, -0.029297582805156708, -0.011229176074266434, -0.0019563811365514994, -0.07089760154485703, ...
7
Invertible Image Rescaling
[ "Mingqing Xiao", "Shuxin Zheng", "Chang Liu", "Yaolong Wang", "Di He", "Guolin Ke", "Jiang Bian", "Zhouchen Lin", "Tie-Yan Liu" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/529_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460120.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460120-supp.pdf
10.1007/978-3-030-58452-8_8
2005.05650
title_snapshot
High-resolution digital images are usually downscaled to fit various display screens or save the cost of storage and bandwidth, meanwhile the post-upscaling is adpoted to recover the original resolutions or the details in the zoom-in images. However, typical image downscaling is a non-injective mapping due to the loss ...
[ -0.014204378239810467, -0.03116779215633869, -0.015029028058052063, 0.008814234286546707, 0.06521467119455338, 0.01505848579108715, 0.010575702413916588, -0.004343611653894186, -0.03072548098862171, -0.06359302997589111, 0.02048180252313614, -0.045375075191259384, -0.04590731859207153, 0.0...
8
Synthesize then Compare: Detecting Failures and Anomalies for Semantic Segmentation
[ "Yingda Xia", "Yi Zhang", "Fengze Liu", "Wei Shen", "Alan L. Yuille" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/612_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460137.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460137-supp.pdf
10.1007/978-3-030-58452-8_9
2003.08440
title_snapshot
The ability to detect failures and anomalies are fundamental requirements for building reliable systems for computer vision applications, especially safety-critical applications of semantic segmentation, such as autonomous driving and medical image analysis. In this paper, we systematically study failure and anomaly de...
[ 0.020696667954325676, -0.026584619656205177, 0.004517142660915852, 0.029428865760564804, 0.051241010427474976, 0.03115498647093773, 0.030016988515853882, 0.01605162024497986, -0.030447954311966896, -0.05941654369235039, -0.04652664065361023, 0.013460388407111168, -0.03199043497443199, 0.00...
9
House-GAN: Relational Generative Adversarial Networks for Graph-constrained House Layout Generation
[ "Nelson Nauata", "Kai-Hung Chang", "Chin-Yi Cheng", "Greg Mori", "Yasutaka Furukawa" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/677_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460154.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460154-supp.pdf
10.1007/978-3-030-58452-8_10
2003.06988
title_snapshot
This paper proposes a novel graph-constrained generative adversarial network, whose generator and discriminator are built upon relational architecture. The main idea is to encode the constraint into the graph structure of its relational networks. We have demonstrated the proposed architecture for a new house layout gen...
[ 0.027873018756508827, 0.0017907947767525911, 0.001577843213453889, 0.04692603275179863, 0.02708231657743454, 0.04010239616036415, 0.03487463295459747, 0.006398614961653948, -0.027468325570225716, -0.059153709560632706, -0.037038836628198624, -0.03202670067548752, -0.07267908751964569, 0.00...
10
Crowdsampling the Plenoptic Function
[ "Zhengqi Li", "Wenqi Xian", "Abe Davis", "Noah Snavely" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/736_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460171.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460171-supp.zip
10.1007/978-3-030-58452-8_11
2007.15194
title_snapshot
Many popular tourist landmarks are captured in a multitude of online, public photos. These photos represent a sparse and unstructured sampling of the plenoptic function for a particular scene. In this paper,we present a new approach to novel view synthesis under time-varying illumination from such data. Our approach bu...
[ 0.019382040947675705, -0.014364365488290787, 0.012314463034272194, 0.016968168318271637, 0.017663124948740005, 0.01591375656425953, 0.006785776000469923, 0.019759206101298332, -0.03878060355782509, -0.0450003445148468, -0.0011043359991163015, -0.015064085833728313, -0.08501400053501129, 0....
11
VoxelPose: Towards Multi-Camera 3D Human Pose Estimation in Wild Environment
[ "Hanyue Tu", "Chunyu Wang", "Wenjun Zeng" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/738_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460188.pdf
null
10.1007/978-3-030-58452-8_12
2004.06239
title_snapshot
We present mph{VoxelPose} to estimate $3$D poses of multiple people from multiple camera views. In contrast to the previous efforts which require to establish cross-view correspondence based on noisy and incomplete $2$D pose estimates, mph{VoxelPose} directly operates in the $3$D space therefore avoids making incorre...
[ 0.013642794452607632, -0.004620866384357214, -0.009268234483897686, 0.040656547993421555, 0.0077077955938875675, 0.029742438346147537, 0.05102290213108063, 0.004967563785612583, -0.049171313643455505, -0.05470636114478111, 0.004231621045619249, -0.03629370406270027, -0.10368922352790833, -...
12
End-to-End Object Detection with Transformers
[ "Nicolas Carion", "Francisco Massa", "Gabriel Synnaeve", "Nicolas Usunier", "Alexander Kirillov", "Sergey Zagoruyko" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/832_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460205.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460205-supp.pdf
10.1007/978-3-030-58452-8_13
2005.12872
title_snapshot
We present a new method that views object detection as a direct set prediction. Our approach streamlines the detection pipeline, effectively removing the need for many hand-designed components like a non-maximum suppression procedure or anchor generation that explicitly encode our prior knowledge about the task. The ma...
[ 0.005509396083652973, -0.01876281574368477, -0.0007133599719963968, 0.03067621961236, 0.01108573004603386, 0.016566215083003044, -0.009096288122236729, 0.019849399104714394, -0.025032687932252884, -0.018429486081004143, -0.05897139012813568, 0.04711335897445679, -0.032413456588983536, -0.0...
13
DeepSFM: Structure From Motion Via Deep Bundle Adjustment
[ "Xingkui Wei", "Yinda Zhang", "Zhuwen Li", "Yanwei Fu", "Xiangyang Xue" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/840_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460222.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460222-supp.pdf
10.1007/978-3-030-58452-8_14
1912.09697
title_snapshot
Structure from motion (SfM) is an essential computer vision problem which has not been well handled by deep learning. One of the promising trends is to apply explicit structural constraint, e.g. 3D cost volume, into the network. However, existing methods usually assume accurate camera poses either from GT or other meth...
[ 0.01939897984266281, -0.008824856020510197, 0.011109300889074802, 0.01354349497705698, 0.038795992732048035, 0.05707935988903046, 0.008972091600298882, 0.017606522887945175, -0.021274521946907043, -0.05168262869119644, 0.01470906101167202, -0.02906290628015995, -0.04832258075475693, -0.000...
14
Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetry
[ "Yifan Xu", "Tianqi Fan", "Yi Yuan", "Gurprit Singh" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1044_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460239.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460239-supp.pdf
10.1007/978-3-030-58452-8_15
2007.13393
title_snapshot
Deep implicit field regression methods are effective for 3D reconstruction from single-view images. However, the impact of different sampling patterns on the reconstruction quality is not well-understood. In this work, we first study the effect of point set discrepancy on the network training. Based on Farthest Point S...
[ -0.009721940383315086, -0.01688867248594761, -0.008864952251315117, 0.026785144582390785, 0.02815954200923443, 0.04914301633834839, 0.020134005695581436, 0.010696607641875744, -0.02999114617705345, -0.07532986998558044, 0.00748272193595767, -0.027875738218426704, -0.06671153008937836, 0.02...
15
Segment as Points for Efficient Online Multi-Object Tracking and Segmentation
[ "Zhenbo Xu", "Wei Zhang", "Xiao Tan", "Wei Yang", "Huan Huang", "Shilei Wen", "Errui Ding", "Liusheng Huang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1059_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460256.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460256-supp.zip
10.1007/978-3-030-58452-8_16
2007.01550
title_snapshot
Current multi-object tracking and segmentation (MOTS) methods follow the tracking-by-detection paradigm and adopt convolutions for feature extraction. However, as affected by the inherent receptive field, convolution based feature extraction inevitably mixes up the foreground features and the background features, resul...
[ 0.0010098728816956282, -0.014347090385854244, 0.02609608881175518, 0.04200483858585358, 0.020260877907276154, 0.041802503168582916, 0.007104161195456982, 0.026336945593357086, -0.04717676341533661, -0.05944672226905823, -0.05186228081583977, -0.019332727417349815, -0.07493223994970322, -0....
16
Conditional Convolutions for Instance Segmentation
[ "Zhi Tian", "Chunhua Shen", "Hao Chen" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1105_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460273.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460273-supp.pdf
10.1007/978-3-030-58452-8_17
2003.05664
title_snapshot
We propose a simple yet effective instance segmentation framework, termed CondInst (conditional convolutions for instance segmentation). Top-performing instance segmentation methods such as Mask R-CNN rely on ROI operations (typically ROIPool or ROIAlign) to obtain the final instance masks. In contrast, we propose to s...
[ 0.005851990543305874, -0.04401291906833649, -0.019676581025123596, 0.04524189606308937, 0.0215916708111763, 0.05432296171784401, 0.0036793104372918606, 0.013279777951538563, -0.02214798890054226, -0.010657674632966518, -0.039333853870630264, 0.014298142865300179, -0.05930783972144127, -0.0...
17
MutualNet: Adaptive ConvNet via Mutual Learning from Network Width and Resolution
[ "Taojiannan Yang", "Sijie Zhu", "Chen Chen", "Shen Yan", "Mi Zhang", "Andrew Willis" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1196_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460290.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460290-supp.pdf
10.1007/978-3-030-58452-8_18
1909.12978
title_snapshot
We propose the width-resolution mutual learning method (MutualNet) to train a network that is executable at dynamic resource constraints to achieve adaptive accuracy-efficiency trade-offs at runtime. Our method trains a cohort of sub-networks with different widths using different input resolutions to mutually learn mul...
[ 0.007945368997752666, -0.04512612149119377, -0.011327722109854221, 0.02379857376217842, 0.01333712786436081, 0.03365855664014816, -0.00004529103171080351, 0.006675774697214365, -0.04273142293095589, -0.04222099855542183, -0.007719326298683882, 0.0034673837944865227, -0.05221979320049286, -...
18
Fashionpedia: Ontology, Segmentation, and an Attribute Localization Dataset
[ "Menglin Jia", "Mengyun Shi", "Mikhail Sirotenko", "Yin Cui", "Claire Cardie", "Bharath Hariharan", "Hartwig Adam", "Serge Belongie" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1203_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460307.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460307-supp.pdf
10.1007/978-3-030-58452-8_19
2004.12276
title_snapshot
Segmentation, and an Attribute Localization Dataset","In this work, we focus on the task of instance segmentation with attribute localization. This unifies instance segmentation (detect and segment each object instance) and visual categorization of fine-grained attributes (classify one or multiple attributes). The prop...
[ 0.02153296209871769, -0.03283599764108658, -0.002618498634546995, 0.050258658826351166, 0.05166392773389816, 0.02582792565226555, 0.01896500028669834, -0.0011686431244015694, -0.01567087508738041, -0.013730811886489391, -0.0800890177488327, 0.004240235313773155, -0.05601426959037781, -0.00...
19
Privacy Preserving Structure-from-Motion
[ "Marcel Geppert", "Viktor Larsson", "Pablo Speciale", "Johannes L. Schönberger", "Marc Pollefeys" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1273_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460324.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460324-supp.zip
10.1007/978-3-030-58452-8_20
null
null
Over the last years, visual localization and mapping solutions have been adopted by an increasing number of mixed reality and robotics systems. The recent trend towards cloud-based localization and mapping systems has raised significant privacy concerns. These are mainly grounded by the fact that these services require...
[ 0.04246581718325615, 0.011407707817852497, 0.004766422789543867, 0.05719160661101341, 0.031241267919540405, 0.05324038490653038, 0.030718663707375526, 0.013504336588084698, -0.047163888812065125, -0.03331691026687622, -0.018672507256269455, -0.071085125207901, -0.05512348189949989, -0.0133...
20
Rewriting a Deep Generative Model
[ "David Bau", "Steven Liu", "Tongzhou Wang", "Jun-Yan Zhu", "Antonio Torralba" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1326_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460341.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460341-supp.zip
10.1007/978-3-030-58452-8_21
2007.15646
title_snapshot
A deep generative model such as a GAN learns to model a rich set of semantic and physical rules about the target distribution, but up to now, it has been obscure how such rules are encoded in the network, or how a rule could be changed. In this paper, we introduce a new problem setting: manipulation of specific rules e...
[ -0.010652424767613411, -0.004316267557442188, -0.01892722025513649, 0.0430046021938324, 0.04776027053594589, 0.008439240977168083, -0.004637741483747959, 0.015487630851566792, -0.0008952861535362899, -0.04044865071773529, -0.007839680649340153, 0.008161102421581745, -0.05821748077869415, 0...
21
Compare and Reweight: Distinctive Image Captioning Using Similar Images Sets
[ "Jiuniu Wang", "Wenjia Xu", "Qingzhong Wang", "Antoni B. Chan" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1417_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460358.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460358-supp.pdf
10.1007/978-3-030-58452-8_22
2007.06877
title_snapshot
A wide range of image captioning models has been developed, achieving significant improvement based on popular metrics, such as BLEU, CIDEr, and SPICE. However, although the generated captions can accurately describe the image, they are generic for similar images and lack distinctiveness, i.e., cannot properly describe...
[ -0.0074324836023151875, -0.039621878415346146, -0.010214493609964848, 0.051661234349012375, 0.03527909144759178, -0.005309537518769503, 0.015415995381772518, 0.017215846106410027, -0.03528214246034622, -0.04934947192668915, -0.045719392597675323, 0.007374953478574753, -0.06893524527549744, ...
22
Long-term Human Motion Prediction with Scene Context
[ "Zhe Cao", "Hang Gao", "Karttikeya Mangalam", "Qi-Zhi Cai", "Minh Vo", "Jitendra Malik" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1448_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460375.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460375-supp.zip
10.1007/978-3-030-58452-8_23
2007.03672
title_snapshot
Human movement is goal-directed and influenced by the spatial layout of the objects in the scene. To plan future human motion, it is crucial to perceive the environment -- imagine how hard it is to navigate a new room with lights off. Existing works on predicting human motion do not pay attention to the scene context a...
[ 0.003543344559147954, -0.018430709838867188, 0.006407133303582668, 0.0006422886508516967, 0.04811544343829155, 0.010052635334432125, 0.03421105816960335, 0.023006808012723923, -0.06244273856282234, -0.039398472756147385, -0.07180067896842957, -0.029084885492920876, -0.06555099040269852, -0...
23
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
[ "Ben Mildenhall", "Pratul P. Srinivasan", "Matthew Tancik", "Jonathan T. Barron", "Ravi Ramamoorthi", "Ren Ng" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1473_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460392.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460392-supp.pdf
10.1007/978-3-030-58452-8_24
2003.08934
title_snapshot
We present a method that achieves state-of-the-art results for synthesizing novel views of complex scenes by optimizing an underlying continuous volumetric scene function using a sparse set of input views. Our algorithm represents a scene using a fully-connected (non-convolutional) deep network, whose input is a single...
[ 0.02849915623664856, 0.0037347176112234592, 0.009589292109012604, 0.02218504808843136, 0.03395231068134308, 0.026942148804664612, -0.013003368861973286, -0.0014566576573997736, -0.037872109562158585, -0.05073511227965355, -0.02898559533059597, -0.002067307708784938, -0.0520847886800766, 0....
24
ReferIt3D: Neural Listeners for Fine-Grained 3D Object Identification in Real-World Scenes
[ "Panos Achlioptas", "Ahmed Abdelreheem", "Fei Xia", "Mohamed Elhoseiny", "Leonidas Guibas" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1501_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460409.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460409-supp.pdf
10.1007/978-3-030-58452-8_25
null
null
In this work we study the problem of using referential language to identify common objects in real-world 3D scenes. We focus on a challenging setup where the referred object belongs to a extit{fine-grained} object class and the underlying scene contains extit{multiple} object instances of that class. Due to the scarcit...
[ -0.011465304531157017, 0.02326657809317112, 0.03592320904135704, 0.005178786348551512, 0.011675947345793247, 0.07497173547744751, 0.008957385085523129, 0.03018050082027912, -0.04490036144852638, -0.04713428020477295, -0.05126800015568733, 0.04207904264330864, -0.04465328902006149, 0.038356...
25
MatryODShka: Real-time 6DoF Video View Synthesis using Multi-Sphere Images
[ "Benjamin Attal", "Selena Ling", "Aaron Gokaslan", "Christian Richardt", "James Tompkin" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1737_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460426.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460426-supp.zip
10.1007/978-3-030-58452-8_26
2008.06534
title_snapshot
We introduce a method to convert stereo 360 (omnidirectional stereo) imagery into a layered, multi-sphere image representation for six degree-of-freedom (6DoF) rendering. Stereo 360 imagery can be captured from multi-camera systems for virtual reality (VR) rendering, but lacks motion parallax and correct-in-all-directi...
[ 0.047831494361162186, 0.0621471032500267, 0.00817530881613493, 0.039282139390707016, -0.002755046123638749, 0.03551400080323219, 0.034097470343112946, 0.037545397877693176, -0.04269289970397949, -0.04157758876681328, -0.03195034712553024, -0.017454972490668297, -0.053518299013376236, 0.036...
26
Learning and Aggregating Deep Local Descriptors for Instance-level Recognition
[ "Giorgos Tolias", "Tomas Jenicek", "Ondřej Chum" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1793_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460443.pdf
null
10.1007/978-3-030-58452-8_27
2007.13172
title_snapshot
We propose an efficient method to learn deep local descriptors for instance-level recognition. The training only requires examples of positive and negative image pairs and is performed as metric learning of sum-pooled global image descriptors. At inference, the local descriptors are provided by the activations of inter...
[ 0.00033692282158881426, -0.026452364400029182, 0.02668318711221218, 0.059720516204833984, 0.02543494664132595, 0.0471048578619957, -0.004965465981513262, -0.01134384423494339, -0.010285426862537861, -0.02233137935400009, -0.016769802197813988, -0.0038414262235164642, -0.0889899879693985, 0...
27
A Consistently Fast and Globally Optimal Solution to the Perspective-n-Point Problem
[ "George Terzakis", "Manolis Lourakis" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/1969_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460460.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460460-supp.pdf
10.1007/978-3-030-58452-8_28
null
null
An approach for estimating the pose of a camera given a set of 3D points and their corresponding 2D image projections is presented. It formulates the problem as a non-linear quadratic program and identifies regions in the parameter space that contain unique minima with guarantees that at least one of them will be the g...
[ -0.00898439809679985, 0.015373867005109787, 0.00725567014887929, 0.023196827620267868, 0.024964991956949234, 0.06772913038730621, -0.003558660391718149, 0.00930958054959774, -0.05353496968746185, -0.034090228378772736, -0.03643350675702095, -0.004000838380306959, -0.06979042291641235, -0.0...
28
Learn to Recover Visible Color for Video Surveillance in a Day
[ "Guangming Wu", "Yinqiang Zheng", "Zhiling Guo", "Zekun Cai", "Xiaodan Shi", "Xin Ding", "Yifei Huang", "Yimin Guo", "Ryosuke Shibasaki" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2096_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460477.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460477-supp.pdf
10.1007/978-3-030-58452-8_29
null
null
In silicon sensors, the interference between visible and near-infrared (NIR) signals is a crucial problem. For all-day video surveillance, commercial camera systems usually adopt auxiliary NIR cut filter and NIR LED illumination to selectively block or enhance NIR signal according to the surrounding light conditions. T...
[ 0.03801388666033745, -0.015138130635023117, -0.01781448721885681, 0.05137418583035469, 0.03479049354791641, 0.012903536669909954, 0.022996604442596436, 0.009185492992401123, -0.03920908272266388, -0.06603577733039856, -0.02063284069299698, 0.004439599812030792, -0.07579894363880157, 0.0088...
29
Deep Fashion3D: A Dataset and Benchmark for 3D Garment Reconstruction from Single Images
[ "Heming Zhu", "Yu Cao", "Hang Jin", "Weikai Chen", "Dong Du", "Zhangye Wang", "Shuguang Cui", "Xiaoguang Han" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2149_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460494.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460494-supp.pdf
10.1007/978-3-030-58452-8_30
2003.12753
title_snapshot
High-fidelity clothing reconstruction is the key to achieving photorealism in a wide range of applications including human digitization, virtual try-on, etc. Recent advances in learning-based approaches have accomplished unprecedented accuracy in recovering unclothed human shape and pose from single images, thanks to t...
[ 0.03232423588633537, -0.033442333340644836, -0.029587801545858383, 0.03423308581113815, 0.0628158375620842, 0.05403852090239525, 0.025498511269688606, -0.002067970810458064, -0.01106457319110632, -0.07185686379671097, -0.03246556967496872, -0.029073204845190048, -0.047574419528245926, 0.00...
30
Spatially Adaptive Inference with Stochastic Feature Sampling and Interpolation
[ "Zhenda Xie", "Zheng Zhang", "Xizhou Zhu", "Gao Huang", "Stephen Lin" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2193_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460511.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460511-supp.pdf
10.1007/978-3-030-58452-8_31
2003.08866
title_snapshot
In the feature maps of CNNs, there commonly exists considerable spatial redundancy that leads to much repetitive processing. Towards reducing this superfluous computation, we propose to compute features only at sparsely sampled locations, which are probabilistically chosen according to activation responses, and then de...
[ 0.019548729062080383, -0.012812326662242413, -0.003114769235253334, 0.02731473743915558, 0.04250825569033623, 0.04947260767221451, 0.022630874067544937, -0.002410808578133583, -0.02857685089111328, -0.054794300347566605, -0.008650826290249825, -0.046932876110076904, -0.0531751848757267, -0...
31
BorderDet: Border Feature for Dense Object Detection
[ "Han Qiu", "Yuchen Ma", "Zeming Li", "Songtao Liu", "Jian Sun" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2211_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460528.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460528-supp.pdf
10.1007/978-3-030-58452-8_32
2007.11056
title_snapshot
Dense object detectors rely on the sliding-window paradigm that predicts the object over a regular grid of image. Meanwhile, the feature maps on the point of the grid are adopted to generate the bounding box predictions. The point feature is convenient to use but may lack the explicit border information for accurate lo...
[ -0.027680015191435814, -0.007532112300395966, 0.016341105103492737, 0.022956354543566704, 0.030255699530243874, 0.01959250681102276, 0.008654278703033924, -0.012429730966687202, -0.036822546273469925, -0.040904588997364044, -0.032851554453372955, -0.032743655145168304, -0.05689560994505882, ...
32
Regularization with Latent Space Virtual Adversarial Training
[ "Genki Osada", "Budrul Ahsan", "Revoti Prasad Bora", "Takashi Nishide" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2258_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460545.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460545-supp.zip
10.1007/978-3-030-58452-8_33
2011.13181
title_snapshot
Virtual Adversarial Training (VAT) has shown impressive results among recently developed regularization methods called consistency regularization. VAT utilizes adversarial samples, generated by injecting perturbation in the input space, for training and thereby enhances the generalization ability of a classifier. Howev...
[ 0.05352457985281944, -0.03948551416397095, -0.02906201221048832, 0.041680965572595596, 0.03854012116789818, 0.017024705186486244, 0.03301563486456871, -0.0158721674233675, -0.024719495326280594, -0.06848463416099548, -0.028509333729743958, -0.017883004620671272, -0.06868802756071091, 0.007...
33
Du²Net: Learning Depth Estimation from Dual-Cameras and Dual-Pixels
[ "Yinda Zhang", "Neal Wadhwa", "Sergio Orts-Escolano", "Christian Häne", "Sean Fanello", "Rahul Garg" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2263_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460562.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460562-supp.zip
10.1007/978-3-030-58452-8_34
2003.14299
title_judge
Computational stereo has reached a high level of accuracy, but degrades in the presence of occlusions, repeated textures, and correspondence errors along edges. We present a novel approach based on neural networks for depth estimation that combines stereo from dual cameras with stereo from a dual-pixel sensor, which is...
[ 0.010845154523849487, 0.014957987703382969, -0.031819701194763184, 0.026845594868063927, 0.028969526290893555, 0.04822888597846031, 0.0171721912920475, 0.008126124739646912, -0.012439272366464138, -0.054084956645965576, 0.015646766871213913, -0.009109602309763432, -0.04321024939417839, 0.0...
34
Model-Agnostic Boundary-Adversarial Sampling for Test-Time Generalization in Few-Shot learning
[ "Jaekyeom Kim", "Hyoungseok Kim", "Gunhee Kim" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2307_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460579.pdf
null
10.1007/978-3-030-58452-8_35
null
null
Few-shot learning is an important research problem that tackles one of the greatest challenges of machine learning: learning a new task from a limited amount of labeled data. We propose a model-agnostic method that improves the test-time performance of any few-shot learning models with no additional training, and thus ...
[ 0.006552661769092083, -0.0428960882127285, 0.00394670944660902, 0.05227940157055855, 0.021506454795598984, -0.0094827339053154, 0.0423705130815506, -0.032172542065382004, -0.010996929369866848, -0.010994325391948223, 0.003315064124763012, 0.0022059702314436436, -0.07377061247825623, 0.0039...
35
Targeted Attack for Deep Hashing based Retrieval
[ "Jiawang Bai", "Bin Chen", "Yiming Li", "Dongxian Wu", "Weiwei Guo", "Shu-Tao Xia", "En-Hui Yang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2463_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460596.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460596-supp.pdf
10.1007/978-3-030-58452-8_36
2004.07955
title_snapshot
The deep hashing based retrieval method is widely adopted in large-scale image and video retrieval. However, there is little investigation on its security. In this paper, we propose a novel method, dubbed deep hashing targeted attack (DHTA), to study the targeted attack on such retrieval. Specifically, we first formula...
[ -0.007478252984583378, -0.015931040048599243, -0.0011301562190055847, 0.05990775302052498, 0.02746778540313244, 0.0012047559721395373, 0.025975190103054047, -0.0065781655721366405, -0.018855290487408638, -0.03814128413796425, -0.019886191934347153, -0.026707081124186516, -0.0486878901720047,...
36
Gradient Centralization: A New Optimization Technique for Deep Neural Networks
[ "Hongwei Yong", "Jianqiang Huang", "Xiansheng Hua", "Lei Zhang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2471_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460613.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460613-supp.pdf
10.1007/978-3-030-58452-8_37
2004.01461
title_snapshot
Optimization techniques are of great importance to effectively and efficiently train a deep neural network (DNN). It has been shown that using the first and second order statistics (e.g., mean and variance) to perform Z-score standardization on network activations or weight vectors, such as batch normalization (BN) and wei...
[ -0.02004399709403515, -0.0237733107060194, -0.00400214409455657, 0.02397122234106064, 0.027954036369919777, 0.05284338444471359, 0.005851049907505512, 0.0036696235183626413, -0.025956863537430763, -0.049174487590789795, -0.000010468434084032197, -0.023907972499728203, -0.03807393088936806, ...
37
Content-Aware Unsupervised Deep Homography Estimation
[ "Jirong Zhang", "Chuan Wang", "Shuaicheng Liu", "Lanpeng Jia", "Nianjin Ye", "Jue Wang", "Ji Zhou", "Jian Sun" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2503_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460630.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460630-supp.zip
10.1007/978-3-030-58452-8_38
1909.05983
title_snapshot
Homography estimation is a basic image alignment method in many applications. It is usually done by extracting and matching sparse feature points, which are error-prone in low-light and low-texture images. On the other hand, previous deep homography approaches use either synthetic images for supervised learning or aeri...
[ 0.04562174901366234, -0.00043167846160940826, 0.0027839348185807467, 0.02276650071144104, 0.02354654110968113, 0.02490757405757904, 0.01932549476623535, 0.011547669768333435, -0.022114595398306847, -0.05192176625132561, -0.016115710139274597, -0.011311898939311504, -0.08059066534042358, -0...
38
Multi-View Optimization of Local Feature Geometry
[ "Mihai Dusmanu", "Johannes L. Schönberger", "Marc Pollefeys" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2556_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460647.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460647-supp.pdf
10.1007/978-3-030-58452-8_39
2003.08348
title_snapshot
In this work, we address the problem of refining the geometry of local image features from multiple views without known scene or camera geometry. Current approaches to local feature detection are inherently limited in their keypoint localization accuracy because they only operate on a single view. This limitation has a...
[ -0.002355381613597274, 0.025198210030794144, 0.018289318308234215, 0.04784511774778366, 0.034748829901218414, 0.048224806785583496, 0.013408957980573177, -0.012289690785109997, -0.038432635366916656, -0.04289286956191063, -0.026832010596990585, -0.02037559263408184, -0.062400851398706436, ...
39
The Phong Surface: Efficient 3D Model Fitting using Lifted Optimization
[ "Jingjing Shen", "Thomas J. Cashman", "Qi Ye", "Tim Hutton", "Toby Sharp", "Federica Bogo", "Andrew Fitzgibbon", "Jamie Shotton" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2597_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460664.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460664-supp.pdf
10.1007/978-3-030-58452-8_40
2007.04940
title_snapshot
Realtime perceptual and interaction capabilities in mixed reality require a range of 3D tracking problems to be solved at low latency on resource-constrained hardware such as head-mounted devices. Indeed, for devices such as HoloLens 2 where the CPU and GPU are left available for applications, multiple tracking subsyst...
[ -0.009294118732213974, 0.06926373392343521, 0.0011777891777455807, -0.004479424562305212, 0.03920621797442436, 0.04178033024072647, 0.0004204912984278053, 0.05138752609491348, -0.03813298046588898, -0.05914295092225075, -0.02193584479391575, -0.00867551937699318, -0.0721038281917572, -0.01...
40
Forecasting Human-Object Interaction: Joint Prediction of Motor Attention and Actions in First Person Video
[ "Miao Liu", "Siyu Tang", "Yin Li", "James M. Rehg" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2641_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460681.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460681-supp.pdf
10.1007/978-3-030-58452-8_41
1911.10967
title_snapshot
We address the challenging task of anticipating human-object interaction in first person videos. Most existing methods either ignore how the camera wearer interacts with objects, or simply considers body motion as a separate modality. In contrast, we observe that the intentional hand movement reveals critical informati...
[ 0.0013391257962211967, 0.001040615257807076, -0.0067962114699184895, -0.015629859641194344, 0.04409932345151901, 0.012672049924731255, 0.04039020836353302, 0.022287271916866302, -0.044545188546180725, -0.04135788604617119, -0.028881700709462166, -0.0035709149669855833, -0.06666628271341324, ...
41
Learning Stereo from Single Images
[ "Jamie Watson", "Oisin Mac Aodha", "Daniyar Turmukhambetov", "Gabriel J. Brostow", "Michael Firman" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2683_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460698.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460698-supp.pdf
10.1007/978-3-030-58452-8_42
2008.01484
title_snapshot
Supervised deep networks are among the best methods for finding correspondences in stereo image pairs. Like all supervised approaches, these networks require ground truth data during training. However, collecting large quantities of accurate dense correspondence data is very challenging. We propose that it is unnecessa...
[ 0.03929703310132027, 0.024824120104312897, -0.029109397903084755, 0.0647481232881546, 0.02225724793970585, 0.05206670984625816, 0.007201933767646551, 0.019421206787228584, 0.0034091584384441376, -0.04078459367156029, -0.012965920381247997, 0.011801853775978088, -0.0747891291975975, 0.01771...
42
Prototype Rectification for Few-Shot Learning
[ "Jinlu Liu", "Liang Song", "Yongqiang Qin" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2748_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460715.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460715-supp.pdf
10.1007/978-3-030-58452-8_43
1911.10713
title_snapshot
Few-shot learning requires to recognize novel classes with scarce labeled data. Prototypical network is useful in existing researches, however, training on narrow-size distribution of scarce data usually tends to get biased prototypes. In this paper, we figure out two key influencing factors of the process: the intra-c...
[ 0.005506567656993866, -0.03883594274520874, -0.034289855509996414, 0.03543783724308014, 0.02469523437321186, 0.021201863884925842, 0.04351508989930153, -0.020295677706599236, -0.013237728737294674, -0.03985503688454628, -0.00017017117352224886, 0.003403406823053956, -0.055555615574121475, ...
43
Learning Feature Descriptors using Camera Pose Supervision
[ "Qianqian Wang", "Xiaowei Zhou", "Bharath Hariharan", "Noah Snavely" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2784_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460732.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460732-supp.pdf
10.1007/978-3-030-58452-8_44
2004.13324
title_snapshot
Recent research on learned visual descriptors has shown promising improvements in correspondence estimation, a key component of many 3D vision tasks. However, existing descriptor learning frameworks typically require ground-truth correspondences between feature points for training, which are challenging to acquire at s...
[ 0.016538510099053383, -0.006040777079761028, 0.005397930275648832, 0.031745847314596176, 0.025039713829755783, 0.03161703050136566, -0.003802013583481312, -0.002173319924622774, -0.016167722642421722, -0.011872461996972561, -0.03114880993962288, -0.006359238177537918, -0.09283397346735, 0....
44
Semantic Flow for Fast and Accurate Scene Parsing
[ "Xiangtai Li", "Ansheng You", "Zhen Zhu", "Houlong Zhao", "Maoke Yang", "Kuiyuan Yang", "Shaohua Tan", "Yunhai Tong" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2785_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460749.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460749-supp.pdf
10.1007/978-3-030-58452-8_45
2002.10120
title_snapshot
In this paper, we focus on designing effective method for fast and accurate scene parsing. A common practice to improve the performance is to attain high resolution feature maps with strong semantic representation. Two strategies are widely used---atrous convolutions and feature pyramid fusion, are either computation i...
[ 0.024233244359493256, -0.014532085508108139, 0.0255154836922884, 0.02905196323990822, 0.01062504481524229, 0.04062869772315025, 0.011607164517045021, 0.02065179869532585, -0.009514417499303818, -0.03836587816476822, -0.02633194997906685, -0.02254141867160797, -0.07092295587062836, 0.000036...
45
Appearance Consensus Driven Self-Supervised Human Mesh Recovery
[ "Jogendra Nath Kundu", "Mugalodi Rakesh", "Varun Jampani", "Rahul Mysore Venkatesh", "R. Venkatesh Babu" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2788_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460766.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123460766-supp.pdf
10.1007/978-3-030-58452-8_46
2008.01341
title_snapshot
We present a self-supervised human mesh recovery framework to infer human pose and shape from monocular images in the absence of any paired supervision. Recent advances have shifted the interest towards directly regressing parameters of a parametric human model by supervising them on large-scale, images with 2D landmar...
[ 0.025495881214737892, -0.013247589580714703, -0.0021438896656036377, 0.028348635882139206, 0.03490215912461281, 0.0352700911462307, 0.018806610256433487, 0.0041369060054421425, -0.05458296835422516, -0.07224980741739273, -0.024323325604200363, -0.029447365552186966, -0.10229208320379257, -...
46
Diffraction Line Imaging
[ "Mark Sheinin", "Dinesh N. Reddy", "Matthew O’Toole", "Srinivasa G. Narasimhan" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2825_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470001.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470001-supp.zip
10.1007/978-3-030-58536-5_1
null
null
We present a novel computational imaging principle that combines diffractive optics with line (1D) sensing. When light passes through a diffraction grating, it disperses as a function of wavelength. We exploit this principle to recover 2D and even 3D positions from only line images. We derive a detailed image formation...
[ 0.003657485591247678, 0.04053284227848053, 0.008366156369447708, 0.01723323203623295, 0.06020963564515114, 0.030913706868886948, 0.019037943333387375, 0.019336719065904617, -0.027616063132882118, -0.060441527515649796, 0.01774718053638935, -0.03148042410612106, -0.044839583337306976, 0.002...
47
Aligning and Projecting Images to Class-conditional Generative Networks
[ "Minyoung Huh", "Richard Zhang", "Jun-Yan Zhu", "Sylvain Paris", "Aaron Hertzmann" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2834_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470018.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470018-supp.pdf
10.1007/978-3-030-58536-5_2
2005.01703
title_judge
We present a method for projecting an input image into the space of a class-conditional generative neural network. We propose a method that optimizes for transformation to counteract the model biases in generative neural networks. Specifically, we demonstrate that one can solve for image translation, scale, and global ...
[ 0.003146548056975007, -0.011920684017241001, -0.017416276037693024, 0.05214978754520416, 0.01934952661395073, 0.04366149753332138, 0.0006779805989935994, -0.00848286785185337, -0.020888198167085648, -0.04031497612595558, -0.035903967916965485, 0.004360812250524759, -0.070293128490448, 0.00...
48
Suppress and Balance: A Simple Gated Network for Salient Object Detection
[ "Xiaoqi Zhao", "Youwei Pang", "Lihe Zhang", "Huchuan Lu", "Lei Zhang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2852_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470035.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470035-supp.pdf
10.1007/978-3-030-58536-5_3
2007.08074
title_snapshot
Most salient object detection approaches use U-Net or feature pyramid networks (FPN) as their basic structures. These methods ignore two key problems when the encoder exchanges information with the decoder: one is the lack of interference control between them, the other is without considering the disparity of the contr...
[ 0.0016563261160627007, -0.004410700872540474, 0.02681039460003376, 0.025604568421840668, 0.02029300294816494, 0.02396412566304207, 0.009341545403003693, -0.026265660300850868, -0.027835708111524582, -0.06080613285303116, 0.01370159536600113, -0.016523024067282677, -0.049567535519599915, -0...
49
Visual Memorability for Robotic Interestingness via Unsupervised Online Learning
[ "Chen Wang", "Wenshan Wang", "Yuheng Qiu", "Yafei Hu", "Sebastian Scherer" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2904_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470052.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470052-supp.pdf
10.1007/978-3-030-58536-5_4
2005.08829
title_snapshot
In this paper, we explore the problem of interesting scene prediction for mobile robots. This area is currently underexplored but is crucial for many practical applications such as autonomous exploration and decision making. Inspired by industrial demands, we first propose a novel translation-invariant visual memory fo...
[ -0.004966624081134796, 0.001193830743432045, 0.001727947499603033, 0.026191746816039085, 0.047202013432979584, -0.0061482894234359264, -0.0014678729930892587, 0.017710063606500626, -0.048179447650909424, -0.03424571454524994, -0.0317201167345047, 0.002697810996323824, -0.0595451258122921, ...
50
Post-Training Piecewise Linear Quantization for Deep Neural Networks
[ "Jun Fang", "Ali Shafiee", "Hamzah Abdel-Aziz", "David Thorsley", "Georgios Georgiadis", "Joseph H. Hassoun" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2949_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470069.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470069-supp.pdf
10.1007/978-3-030-58536-5_5
2002.00104
title_snapshot
Quantization plays an important role in the energy-efficient deployment of Deep Neural Networks (DNNs) on resource-limited devices. Post-training quantization is highly desirable since it does not require retraining or access to the full training dataset. The well-established uniform scheme for post-training quantizati...
[ -0.01841643825173378, -0.02667560800909996, -0.032846253365278244, 0.029607558622956276, 0.04692188650369644, 0.052967727184295654, 0.004250687081366777, -0.02149687334895134, -0.038730867207050323, -0.039569586515426636, -0.024569641798734665, -0.03051520325243473, -0.05816671997308731, 0...
51
Joint Disentangling and Adaptation for Cross-Domain Person Re-Identification
[ "Yang Zou", "Xiaodong Yang", "Zhiding Yu", "B.V.K. Vijaya Kumar", "Jan Kautz" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2974_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470086.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470086-supp.pdf
10.1007/978-3-030-58536-5_6
2007.10315
title_snapshot
Although a significant progress has been witnessed in supervised person re-identification (re-id), it remains challenging to generalize re-id models to new domains due to the huge domain gaps. Recently, there has been a growing interest in using unsupervised domain adaptation to address this scalability issue. Existing...
[ 0.001985246315598488, -0.02285602129995823, -0.004068007692694664, 0.04859811067581177, 0.059191975742578506, 0.0005081435083411634, 0.02984403446316719, -0.02345823124051094, -0.019231120124459267, -0.05118391290307045, -0.018200507387518883, -0.00584773812443018, -0.1004311591386795, -0....
52
In-Home Daily-Life Captioning Using Radio Signals
[ "Lijie Fan", "Tianhong Li", "Yuan Yuan", "Dina Katabi" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2978_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470103.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470103-supp.pdf
10.1007/978-3-030-58536-5_7
2008.10966
title_snapshot
This paper aims to caption daily life --i.e., to create a textual description of people's activities and interactions with objects in their homes. Addressing this problem requires novel methods beyond traditional video captioning, as most people would have privacy concerns about deploying cameras throughout their homes...
[ 0.014758484438061714, 0.005306445527821779, 0.042017921805381775, 0.00010071127326227725, 0.03988119959831238, -0.034875813871622086, 0.02545146457850933, 0.00758404703810811, -0.026003418490290642, -0.025069745257496834, -0.04415947571396828, -0.00001717951090540737, -0.06919927150011063, ...
53
Self-Challenging Improves Cross-Domain Generalization
[ "Zeyi Huang", "Haohan Wang", "Eric P. Xing", "Dong Huang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3018_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470120.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470120-supp.pdf
10.1007/978-3-030-58536-5_8
2007.02454
title_snapshot
Convolutional Neural Networks (CNN) conduct image classification by activating dominant features that correlated with labels. When the training and testing data are under similar distributions, their dominant features are similar, leading to decent test performance. The performance is nonetheless unmet when tested with...
[ -0.006730780936777592, -0.03808055818080902, -0.003978211898356676, 0.03691290318965912, 0.030814778059720993, 0.01580573432147503, 0.02057046815752983, -0.020207533612847328, -0.019445819780230522, -0.021933529525995255, -0.020892079919576645, -0.005462217144668102, -0.046711526811122894, ...
54
A Competence-aware Curriculum for Visual Concepts Learning via Question Answering
[ "Qing Li", "Siyuan Huang", "Yining Hong", "Song-Chun Zhu" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3029_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470137.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470137-supp.pdf
10.1007/978-3-030-58536-5_9
2007.01499
title_snapshot
Humans can progressively learn visual concepts from easy to hard questions. To mimic this efficient learning ability, we propose a competence-aware curriculum for visual concept learning in a question-answering manner. Specifically, we design a neural-symbolic concept learner for learning the visual concepts and a mult...
[ -0.007354125380516052, -0.01827956736087799, 0.00949178822338581, 0.05468899756669998, 0.04587022215127945, 0.014916031621396542, 0.015241885557770729, 0.018632924184203148, -0.03518228232860565, 0.004219829570502043, -0.03882654383778572, 0.0401591882109642, -0.03294287249445915, 0.016665...
55
Multitask Learning Strengthens Adversarial Robustness
[ "Chengzhi Mao", "Amogh Gupta", "Vikram Nitin", "Baishakhi Ray", "Shuran Song", "Junfeng Yang", "Carl Vondrick" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3047_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470154.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470154-supp.pdf
10.1007/978-3-030-58536-5_10
2007.07236
title_snapshot
Although deep networks achieve strong accuracy on a range of computer vision benchmarks, they remain vulnerable to adversarial attacks, where imperceptible input perturbations fool the network. We present both theoretical and empirical analyses that connect the adversarial robustness of a model to the number of tasks t...
[ 0.0036507926415652037, -0.020223813131451607, -0.008891916833817959, 0.050885334610939026, 0.01699623093008995, 0.008698162622749805, 0.04674087464809418, 0.004189711529761553, -0.012654750607907772, -0.06511899828910828, -0.004257930442690849, -0.0001448564580641687, -0.06481704115867615, ...
56
S2DNAS: Transforming Static CNN Model for Dynamic Inference via Neural Architecture Search
[ "Zhihang Yuan", "Bingzhe Wu", "Guangyu Sun", "Zheng Liang", "Shiwan Zhao", "Weichen Bi" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3054_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470171.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470171-supp.pdf
10.1007/978-3-030-58536-5_11
1911.07033
title_snapshot
Recently, dynamic inference has emerged as a promising way to reduce the computational cost of deep convolutional neural networks (CNNs). In contrast to static methods (e.g., weight pruning), dynamic inference adaptively adjusts the inference process according to each input sample, which can considerably reduce the com...
[ -0.0011024826671928167, -0.029712466523051262, -0.030957089737057686, 0.04157419130206108, 0.03476344048976898, 0.04198313504457474, -0.012102755717933178, 0.0038037612102925777, -0.014822988770902157, -0.037173472344875336, 0.029173141345381737, -0.02191864140331745, -0.051079802215099335, ...
57
Improving Deep Video Compression by Resolution-adaptive Flow Coding
[ "Zhihao Hu", "Zhenghao Chen", "Dong Xu", "Guo Lu", "Wanli Ouyang", "Shuhang Gu" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3112_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470188.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470188-supp.zip
10.1007/978-3-030-58536-5_12
2009.05982
title_snapshot
In the learning based video compression approaches, it is an essential issue to compress pixel-level optical flow maps by developing new motion vector (MV) encoders. In this work, we propose a new framework called Resolution-adaptive Flow Coding (RaFC) to effectively compress the flow maps globally and locally, in whic...
[ 0.009396862238645554, -0.003412739373743534, 0.012630744837224483, 0.01692451536655426, 0.06126531586050987, 0.03290081024169922, 0.003392924554646015, -0.008384695276618004, -0.02918544039130211, -0.06740937381982803, -0.004106560256332159, -0.027547966688871384, -0.0140280956402421, 0.01...
58
Motion Capture from Internet Videos
[ "Junting Dong", "Qing Shuai", "Yuanqing Zhang", "Xian Liu", "Xiaowei Zhou", "Hujun Bao" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3158_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470205.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470205-supp.zip
10.1007/978-3-030-58536-5_13
2008.07931
title_snapshot
Recent advances in image-based human pose estimation make it possible to capture 3D human motion from a single RGB video. However, the inherent depth ambiguity and self-occlusion in a single view prohibit the recovery of as high-quality motion as multi-view reconstruction. While multi-view videos are not common, the vi...
[ 0.028401963412761688, -0.02663392946124077, -0.004340194631367922, 0.04224449396133423, 0.048273902386426926, 0.024668671190738678, 0.030511336401104927, 0.012527667917311192, -0.0503719262778759, -0.0417429581284523, -0.026972079649567604, -0.0623638741672039, -0.06933677941560745, -0.031...
59
Appearance-Preserving 3D Convolution for Video-based Person Re-identification
[ "Xinqian Gu", "Hong Chang", "Bingpeng Ma", "Hongkai Zhang", "Xilin Chen" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3183_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470222.pdf
null
10.1007/978-3-030-58536-5_14
2007.08434
title_snapshot
Due to the imperfect person detection results and posture changes, temporal appearance misalignment is unavoidable in video-based person re-identification (ReID). In this case, 3D convolution may destroy the appearance representation of person video clips, thus it is harmful to ReID. To address this problem, we propose...
[ 0.01711246930062771, -0.030296169221401215, 0.01827041432261467, 0.04692321643233299, 0.033434394747018814, 0.042131610214710236, 0.022465383633971214, 0.001940608024597168, -0.024923577904701233, -0.07425343245267868, 0.007288291119039059, -0.010413995943963528, -0.07785169780254364, 0.01...
60
Solving the Blind Perspective-n-Point Problem End-To-End With Robust Differentiable Geometric Optimization
[ "Dylan Campbell", "Liu Liu", "Stephen Gould" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3241_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470239.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470239-supp.zip
10.1007/978-3-030-58536-5_15
2007.14628
title_snapshot
Blind Perspective-n-Point (PnP) is the problem of estimating the position and orientation of a camera relative to a scene, given 2D image points and 3D scene points, without prior knowledge of the 2D-3D correspondences. Solving for pose and correspondences simultaneously is extremely challenging since the search space ...
[ 0.010186259634792805, -0.0077915918081998825, 0.001682138885371387, 0.02677941881120205, 0.010793676599860191, 0.062134165316820145, 0.02373248152434826, 0.010242978110909462, -0.04773804172873497, -0.04227279871702194, -0.03521381691098213, -0.025707781314849854, -0.0568779781460762, -0.0...
61
Exploiting Deep Generative Prior for Versatile Image Restoration and Manipulation
[ "Xingang Pan", "Xiaohang Zhan", "Bo Dai", "Dahua Lin", "Chen Change Loy", "Ping Luo" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3265_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470256.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470256-supp.zip
10.1007/978-3-030-58536-5_16
2003.13659
title_snapshot
Learning a good image prior is a long-term goal for image restoration and manipulation. While existing methods like deep image prior (DIP) capture low-level image statistics, there are still gaps toward an image prior that captures rich image semantics including color, spatial coherence, textures, and high-level concep...
[ 0.017843740060925484, -0.020968565717339516, -0.02692459151148796, 0.07447037100791931, 0.03109852597117424, 0.02494964934885502, 0.012690113857388496, 0.011032403446733952, -0.02300124429166317, -0.09170807152986526, -0.027898624539375305, -0.0055373664945364, -0.04684825986623764, -0.000...
62
Deep Spatial-angular Regularization for Compressive Light Field Reconstruction over Coded Apertures
[ "Mantang Guo", "Junhui Hou", "Jing Jin", "Jie Chen", "Lap-Pui Chau" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3312_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470273.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470273-supp.pdf
10.1007/978-3-030-58536-5_17
2007.11882
title_snapshot
Coded aperture is a promising approach for capturing the 4-D light field (LF), in which the 4-D data are compressively modulated into 2-D coded measurements that are further decoded by reconstruction algorithms. The bottleneck lies in the reconstruction algorithms, resulting in rather limited reconstruction quality. T...
[ 0.01238743681460619, -0.004831204190850258, 0.011809154413640499, 0.004717646632343531, 0.06477060168981552, 0.024702060967683792, -0.000885053479578346, 0.01548638753592968, -0.015448779799044132, -0.06203610450029373, -0.008392083458602428, -0.0176949892193079, -0.024158548563718796, -0....
63
Video-based Remote Physiological Measurement via Cross-verified Feature Disentangling
[ "Xuesong Niu", "Zitong Yu", "Hu Han", "Xiaobai Li", "Shiguang Shan", "Guoying Zhao" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3331_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470290.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470290-supp.pdf
10.1007/978-3-030-58536-5_18
2007.08213
title_snapshot
Remote physiological measurements, e.g., remote photoplethysmography (rPPG) based heart rate (HR), heart rate variability (HRV) and respiration frequency (RF) measuring, are playing more and more important roles under the application scenarios where contact measurement is inconvenient or impossible. Since the amplitude...
[ 0.05693914368748665, 0.009342502802610397, 0.009011579677462578, 0.03013821877539158, 0.05916326120495796, 0.021074969321489334, 0.05031992495059967, -0.0319373682141304, -0.014224485494196415, -0.04263744503259659, 0.008953146636486053, -0.018994344398379326, -0.07319612056016922, -0.0160...
64
Combining Implicit Function Learning and Parametric Models for 3D Human Reconstruction
[ "Bharat Lal Bhatnagar", "Cristian Sminchisescu", "Christian Theobalt", "Gerard Pons-Moll" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3356_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470307.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470307-supp.pdf
10.1007/978-3-030-58536-5_19
2007.11432
title_snapshot
Implicit functions represented as deep learning approximations are powerful for reconstructing 3D surfaces. However, they can only produce static surfaces that are not controllable, which provides limited ability to modify the resulting model by editing its pose or shape parameters.Implicit functions represented as dee...
[ 0.004826878663152456, -0.002267005853354931, -0.04127713665366173, 0.019902892410755157, 0.04140467941761017, 0.06612682342529297, 0.02633458375930786, -0.0020262773614376783, -0.017296073958277702, -0.07063518464565277, -0.034181248396635056, -0.011177955195307732, -0.056575957685709, 0.0...
65
Orientation-aware Vehicle Re-identification with Semantics-guided Part Attention Network
[ "Tsai-Shien Chen", "Chih-Ting Liu", "Chih-Wei Wu", "Shao-Yi Chien" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3376_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470324.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470324-supp.pdf
10.1007/978-3-030-58536-5_20
2008.11423
title_snapshot
Vehicle re-identification (re-ID) focuses on matching images of the same vehicle across different cameras. It is fundamentally challenging because differences between vehicles are sometimes subtle. While several studies incorporate spatial-attention mechanisms to help vehicle re-ID, they often require expensive keypoin...
[ 0.023946911096572876, -0.02361169457435608, 0.019304918125271797, 0.054811764508485794, 0.0352863110601902, 0.04068399965763092, 0.015328601002693176, 0.020564401522278786, -0.0314076729118824, -0.03971649333834648, -0.040817778557538986, 0.003581119468435645, -0.04453146830201149, 0.00161...
66
Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentation
[ "Guolei Sun", "Wenguan Wang", "Jifeng Dai", "Luc Van Gool" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3387_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470341.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470341-supp.pdf
10.1007/978-3-030-58536-5_21
2007.01947
title_snapshot
This paper studies the problem of learning semantic segmentation from image-level supervision only. Current popular solutions leverage object localization maps from classifiers as supervision signals, and struggle to make the localization maps capture more complete object content. Rather than previous efforts that prim...
[ 0.011081025935709476, -0.012744794599711895, 0.009367096237838268, 0.017813431099057198, 0.025895867496728897, 0.025113742798566818, 0.00430089607834816, 0.014709793962538242, -0.013530686497688293, -0.01956331916153431, -0.06199225038290024, -0.010709233582019806, -0.04954548552632332, -0...
67
CoReNet: Coherent 3D Scene Reconstruction from a Single RGB Image
[ "Stefan Popov", "Pablo Bauszat", "Vittorio Ferrari" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3439_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470358.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470358-supp.pdf
10.1007/978-3-030-58536-5_22
2004.12989
title_snapshot
Advances in deep learning techniques have allowed recent work to reconstruct the shape of a single object given only one RBG image as input. Building on common encoder-decoder architectures for this task, we propose three extensions: (1) ray-traced skip connections that propagate local 2D information to the output 3D v...
[ 0.0014885594137012959, -0.02800929546356201, -0.0011284350184723735, 0.02420244738459587, 0.03219246491789818, 0.035293664783239365, 0.0036623699124902487, 0.015309528447687626, -0.04117865860462189, -0.07786409556865692, -0.0061500477604568005, -0.03704133257269859, -0.04768260568380356, ...
68
Layer-wise Conditioning Analysis in Exploring the Learning Dynamics of DNNs
[ "Lei Huang", "Jie Qin", "Li Liu", "Fan Zhu", "Ling Shao" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3482_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470375.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470375-supp.zip
10.1007/978-3-030-58536-5_23
2002.10801
title_snapshot
Conditioning analysis uncovers the landscape of an optimization objective by exploring the spectrum of its curvature matrix. This has been well explored theoretically for linear models. We extend this analysis to deep neural networks (DNNs) in order to investigate their learning dynamics. To this end, we propose layer-...
[ -0.026936890557408333, -0.013558969832956791, 0.0005904634017497301, 0.014780743047595024, 0.03997906669974327, 0.039172206073999405, 0.012824994511902332, -0.022081075236201286, -0.02097189985215664, -0.0342099703848362, -0.004216615576297045, 0.01266911718994379, -0.03654993325471878, -0...
69
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
[ "Zachary Teed", "Jia Deng" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3526_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470392.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470392-supp.pdf
10.1007/978-3-030-58536-5_24
2003.12039
title_snapshot
We introduce Recurrent All-Pairs Field Transforms (RAFT), a new deep network architecture for estimating optical flow. RAFT extracts per-pixel features, builds multi-scale 4D correlation volumes for all pairs of pixels, and iteratively updates a flow field through a recurrent unit that performs lookups on the correlati...
[ 0.020722920075058937, -0.01316385343670845, 0.029644126072525978, 0.010950528085231781, 0.026148568838834763, 0.037499915808439255, 0.004596504848450422, 0.004128108732402325, -0.031202120706439018, -0.04124171659350395, -0.0242975652217865, -0.05038968101143837, -0.052576541900634766, -0....
70
Domain-invariant Stereo Matching Networks
[ "Feihu Zhang", "Xiaojuan Qi", "Ruigang Yang", "Victor Prisacariu", "Benjamin Wah", "Philip Torr" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3528_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470409.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470409-supp.pdf
10.1007/978-3-030-58536-5_25
1911.13287
title_snapshot
State-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes....
[ 0.011919239535927773, 0.012850981205701828, 0.010070684365928173, 0.06134650483727455, 0.04873187467455864, 0.04916287586092949, 0.0050981189124286175, -0.002350317547097802, -0.0066154287196695805, -0.05716003477573395, -0.019219212234020233, -0.010119611397385597, -0.0891670286655426, 0....
71
DeepHandMesh: A Weakly-supervised Deep Encoder-Decoder Framework for High-fidelity Hand Mesh Modeling
[ "Gyeongsik Moon", "Takaaki Shiratori", "Kyoung Mu Lee" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3538_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470426.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470426-supp.zip
10.1007/978-3-030-58536-5_26
2008.08213
title_snapshot
Human hands play a central role in interacting with other people and objects. For realistic replication of such hand motions, high-fidelity hand meshes have to be reconstructed. In this study, we firstly propose DeepHandMesh, a weakly-supervised deep encoder-decoder framework for high-fidelity hand mesh modeling. We de...
[ -0.01566278375685215, -0.006048543378710747, -0.034114740788936615, 0.01295552123337984, 0.042494818568229675, 0.04917062819004059, 0.03501490503549576, -0.003272533416748047, -0.007330028340220451, -0.06978785246610641, 0.015453068539500237, -0.005695488769561052, -0.0665014311671257, 0.0...
72
Content Adaptive and Error Propagation Aware Deep Video Compression
[ "Guo Lu", "Chunlei Cai", "Xiaoyun Zhang", "Li Chen", "Wanli Ouyang", "Dong Xu", "Zhiyong Gao" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3544_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470443.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470443-supp.zip
10.1007/978-3-030-58536-5_27
2003.11282
title_snapshot
Recently, learning based video compression methods attract increasing attention. However, previous works suffer from error propagation, which stems from the accumulation of reconstructed error in inter predictive coding. Meanwhile, previous learning based video codecs are also not adaptive to different video contents. ...
[ 0.0181353110820055, -0.017318755388259888, -0.013784928247332573, 0.04545408859848976, 0.06606099754571915, 0.058276042342185974, 0.013020303100347519, -0.013683428056538105, -0.020687177777290344, -0.047844722867012024, -0.012558397836983204, 0.01217833161354065, -0.0211118645966053, 0.03...
73
Towards Streaming Perception
[ "Mengtian Li", "Yu-Xiong Wang", "Deva Ramanan" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3553_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470460.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470460-supp.pdf
10.1007/978-3-030-58536-5_28
2005.10420
title_snapshot
Embodied perception refers to the ability of an autonomous agent to perceive its environment so that it can (re)act. The responsiveness of the agent is largely governed by latency of its processing pipeline. While past work has studied the algorithmic trade-off between latency and accuracy, there has not been a clear m...
[ 0.013376428745687008, -0.013104597106575966, 0.01041342131793499, 0.032282013446092606, 0.013501642271876335, 0.03299792855978012, 0.011910336092114449, 0.04192923754453659, -0.04160470515489578, -0.05151795968413353, -0.051871974021196365, -0.010710762813687325, -0.06806092709302902, -0.0...
74
Towards Automated Testing and Robustification by Semantic Adversarial Data Generation
[ "Rakshith Shetty", "Mario Fritz", "Bernt Schiele" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3570_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470477.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470477-supp.pdf
10.1007/978-3-030-58536-5_29
null
null
Widespread application of computer vision systems in real world tasks is currently hindered by their unexpected behavior on unseen examples. This occurs due to limitations of empirical testing on finite test sets and lack of systematic methods to identify the breaking points of a trained model. In this work we propose ...
[ 0.012904577888548374, -0.016554849222302437, -0.0072905258275568485, 0.06690067052841187, 0.05309191718697548, 0.021598706021904945, 0.020223278552293777, -0.009627290070056915, -0.029387593269348145, -0.04507627710700035, -0.04190852865576744, 0.009937849827110767, -0.06867995113134384, -...
75
Adversarial Generative Grammars for Human Activity Prediction
[ "AJ Piergiovanni", "Anelia Angelova", "Alexander Toshev", "Michael S. Ryoo" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3582_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470494.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470494-supp.pdf
10.1007/978-3-030-58536-5_30
2008.04888
title_snapshot
In this paper we propose an adversarial generative grammar model for future prediction. The objective is to learn a model that explicitly captures temporal dependencies, providing a capability to forecast multiple, distinct future activities. Our adversarial grammar is designed so that it can learn stochastic productio...
[ 0.02415597438812256, -0.009036378934979439, -0.017261506989598274, 0.02445816434919834, 0.034884899854660034, 0.017465708777308464, 0.031271792948246, 0.011029676534235477, -0.03858311101794243, -0.020742366090416908, -0.03841651231050491, 0.004682000260800123, -0.07557887583971024, -0.000...
76
GDumb: A Simple Approach that Questions Our Progress in Continual Learning
[ "Ameya Prabhu", "Philip H. S. Torr", "Puneet K. Dokania" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3587_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470511.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470511-supp.pdf
10.1007/978-3-030-58536-5_31
null
null
We discuss a general formulation for the Continual Learning (CL) problem for classification---a learning task where a stream provides samples to a learner and the goal of the learner, depending on the samples it receives, is to continually upgrade its knowledge about the old classes and learn new ones. Our formulation ...
[ -0.003151083830744028, -0.06243278831243515, -0.024993199855089188, 0.03908379375934601, 0.04061617702245712, 0.010210483334958553, 0.0021429327316582203, 0.023156093433499336, -0.017634645104408264, -0.0065435622818768024, -0.012089642696082592, 0.019508222118020058, -0.09321361035108566, ...
77
Learning Lane Graph Representations for Motion Forecasting
[ "Ming Liang", "Bin Yang", "Rui Hu", "Yun Chen", "Renjie Liao", "Song Feng", "Raquel Urtasun" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3622_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470528.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470528-supp.zip
10.1007/978-3-030-58536-5_32
2007.13732
title_snapshot
We propose a motion forecasting model that exploits a novel structured map representation as well as actor-map interactions. Instead of encoding vectorized maps as raster images, we construct a lane graph from raw map data to explicitly preserve the map structure. To capture the complex topology and long range dependen...
[ -0.007347735110670328, -0.04384629800915718, 0.04227925091981888, 0.032307371497154236, 0.03293951600790024, 0.027314215898513794, 0.018069539219141006, 0.04456208646297455, -0.04595436155796051, -0.04560357704758644, 0.02089460752904415, -0.010095986537635326, -0.06846405565738678, -0.005...
78
What Matters in Unsupervised Optical Flow
[ "Rico Jonschkowski", "Austin Stone", "Jonathan T. Barron", "Ariel Gordon", "Kurt Konolige", "Anelia Angelova" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3651_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470545.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470545-supp.zip
10.1007/978-3-030-58536-5_33
2006.04902
title_snapshot
We systematically compare and analyze a set of key components in unsupervised optical flow to identify which photometric loss, occlusion handling, and smoothness regularization is most effective. Alongside this investigation we construct a number of novel improvements to unsupervised flow models, such as cost volume no...
[ 0.026578906923532486, -0.02499728836119175, 0.03098476678133011, 0.028191015124320984, 0.03568258509039879, 0.02998065948486328, 0.026941925287246704, 0.0041565923020243645, -0.0269820224493742, -0.06243167445063591, -0.03499670699238777, -0.017093440517783165, -0.07249365746974945, 0.0014...
79
Synthesis and Completion of Facades from Satellite Imagery
[ "Xiaowei Zhang", "Christopher May", "Daniel Aliaga" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3678_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470562.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470562-supp.zip
10.1007/978-3-030-58536-5_34
null
null
Automatic satellite-based reconstruction enables large and widespread creation of urban areas. However, satellite imagery is often noisy and incomplete, and is not suitable for reconstructing detailed building facades. We present a machine learning-based inverse procedural modeling method to automatically create synthe...
[ 0.03134322538971901, -0.034330036491155624, -0.00874400045722723, 0.035483069717884064, 0.06002100557088852, 0.041237395256757736, 0.047339923679828644, 0.029796911403536797, -0.041886407881975174, -0.08009671419858932, -0.05905783176422119, -0.043462395668029785, -0.05613052845001221, -0....
80
Mapillary Planet-Scale Depth Dataset
[ "Manuel López Antequera", "Pau Gargallo", "Markus Hofinger", "Samuel Rota Bulò", "Yubin Kuang", "Peter Kontschieder" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3772_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470579.pdf
null
10.1007/978-3-030-58536-5_35
null
null
Learning-based methods produce remarkable results on single image depth tasks when trained on well-established benchmarks, however, there is a large gap from these benchmarks to real-world performance that is usually obscured by the common practice of fine-tuning on the target dataset. We introduce a new depth dataset ...
[ 0.019484195858240128, -0.028038140386343002, 0.015532923862338066, 0.033381346613168716, 0.04314888268709183, 0.034580543637275696, 0.04031955823302269, 0.01517978310585022, -0.03520704433321953, -0.05382604897022247, -0.005423381458967924, -0.0055524990893900394, -0.06741029769182205, -0....
81
V2VNet: Vehicle-to-Vehicle Communication for Joint Perception and Prediction
[ "Tsun-Hsuan Wang", "Sivabalan Manivasagam", "Ming Liang", "Bin Yang", "Wenyuan Zeng", "Raquel Urtasun" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3838_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470596.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470596-supp.zip
10.1007/978-3-030-58536-5_36
2008.07519
title_snapshot
In this paper, we explore the use of vehicle-to-vehicle (V2V) communication to improve the perception and motion forecasting performance of self-driving vehicles. By intelligently aggregating the information received from multiple nearby vehicles, we can observe the same scene from different viewpoints. This allows us ...
[ 0.025809530168771744, -0.034680526703596115, 0.01949259452521801, 0.040363609790802, 0.01819419674575329, 0.06161389499902725, 0.04944154992699623, 0.01716586947441101, -0.005748179741203785, -0.08249668031930923, -0.0064314547926187515, 0.01702403649687767, -0.06819219142198563, -0.007177...
82
Training Interpretable Convolutional Neural Networks by Differentiating Class-specific Filters
[ "Haoyu Liang", "Zhihao Ouyang", "Yuyuan Zeng", "Hang Su", "Zihao He", "Shu-Tao Xia", "Jun Zhu", "Bo Zhang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3891_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470613.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470613-supp.pdf
10.1007/978-3-030-58536-5_37
2007.08194
title_snapshot
Convolutional neural networks (CNNs) have been successfully used in a range of tasks. However, CNNs are often viewed as ""black-box"" and lack of interpretability. One main reason is due to the filter-class entanglement -- an intricate many-to-many correspondence between filters and classes. Most existing works attempt...
[ 0.016445966437458992, -0.020808441564440727, -0.004347382113337517, 0.029126595705747604, 0.030180824920535088, 0.008352254517376423, 0.007222555577754974, -0.01711556874215603, -0.03926292806863785, -0.05230558291077614, -0.018836164847016335, -0.030036449432373047, -0.08146059513092041, ...
83
EagleEye: Fast Sub-net Evaluation for Efficient Neural Network Pruning
[ "Bailin Li", "Bowen Wu", "Jiang Su", "Guangrun Wang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3948_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470630.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470630-supp.pdf
10.1007/978-3-030-58536-5_38
2007.02491
title_snapshot
Finding out the computational redundant part of a trained Deep Neural Network (DNN) is the key question that pruning algorithms target on. Many algorithms try to predict model performance of the pruned sub-nets by introducing various evaluation methods. But they are either inaccurate or very complicated for general app...
[ 0.0038815424777567387, -0.029531413689255714, -0.017433565109968185, 0.019755279645323753, 0.0427018478512764, 0.07234425097703934, 0.012040350586175919, -0.017271902412176132, -0.02280331403017044, -0.04492327570915222, -0.019017452374100685, -0.00843465980142355, -0.07443058490753174, -0...
84
Intrinsic Point Cloud Interpolation via Dual Latent Space Navigation
[ "Marie-Julie Rakotosaona", "Maks Ovsjanikov" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3975_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470647.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470647-supp.zip
10.1007/978-3-030-58536-5_39
2004.01661
title_snapshot
We present a learning-based method for interpolating and manipulating 3D shapes represented as point clouds, that is explicitly designed to preserve intrinsic shape properties. Our approach is based on constructing a dual encoding space that enables shape synthesis and, at the same time, provides links to the intrinsic...
[ 0.011786234565079212, 0.01289659645408392, -0.01849314384162426, 0.04212264344096184, 0.04188758134841919, 0.06321685016155243, -0.01729629375040531, -0.003191259689629078, -0.02501056343317032, -0.08549077808856964, -0.05170159414410591, -0.040127430111169815, -0.040075644850730896, -0.01...
85
Cross-Domain Cascaded Deep Translation
[ "Oren Katzir", "Dani Lischinski", "Daniel Cohen-Or" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/3976_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470664.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470664-supp.pdf
10.1007/978-3-030-58536-5_40
null
null
In recent years we have witnessed tremendous progress in unpaired image-to-image translation, propelled by the emergence of DNNs and adversarial training strategies. However, most existing methods focus on transfer of style and appearance, rather than on shape translation. The latter task is challenging, due to its int...
[ 0.006790669169276953, -0.023325597867369652, 0.01856345683336258, 0.055430784821510315, 0.038292478770017624, 0.030942747369408607, 0.003541095182299614, 0.003225334221497178, 0.003006170503795147, -0.05297863855957985, -0.01314638927578926, -0.007029140368103981, -0.03839399293065071, -0....
86
“Look Ma, no landmarks!” – Unsupervised, Model-based Dense Face Alignment
[ "Tatsuro Koizumi", "William A. P. Smith" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4043_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470681.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470681-supp.zip
10.1007/978-3-030-58536-5_41
null
null
no landmarks!"" - Unsupervised, model-based dense face alignment","In this paper, we show how to train an image-to-image network to predict dense correspondence between a face image and a 3D morphable model using only the model for supervision. We show that both geometric parameters (shape, pose and camera intrinsics) ...
[ 0.0002889324678108096, 0.00036828595330007374, -0.0030178921297192574, -0.007548417430371046, 0.020122913643717766, 0.03685254603624344, 0.040719423443078995, 0.00517779728397727, -0.0219015683978796, -0.05620608106255531, -0.012432805262506008, 0.022036317735910416, -0.09242270141839981, ...
87
Online Invariance Selection for Local Feature Descriptors
[ "Rémi Pautrat", "Viktor Larsson", "Martin R. Oswald", "Marc Pollefeys" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4158_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470698.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470698-supp.zip
10.1007/978-3-030-58536-5_42
2007.08988
title_snapshot
To be invariant, or not to be invariant: that is the question formulated in this work about local descriptors. A limitation of current feature descriptors is the trade-off between generalization and discriminative power: more invariance means less informative descriptors. We propose to overcome this limitation with a d...
[ 0.015052487142384052, 0.01310728583484888, 0.025118540972471237, 0.026721235364675522, 0.038643576204776764, 0.03841549903154373, 0.026729412376880646, -0.001558502553962171, -0.032382313162088394, -0.0372668020427227, -0.053437408059835434, -0.02712218649685383, -0.0881497859954834, -0.01...
88
Rethinking Image Inpainting via a Mutual Encoder-Decoder with Feature Equalizations
[ "Hongyu Liu", "Bin Jiang", "Yibing Song", "Wei Huang", "Chao Yang" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4179_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470715.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470715-supp.pdf
10.1007/978-3-030-58536-5_43
2007.06929
title_snapshot
Deep encoder-decoder based CNNs have advanced image inpainting methods for hole filling. While existing methods recover structures and textures step-by-step in the hole regions, they typically use two encoder-decoders for separate recovery. The CNN features of each encoder are learned to capture either missing structur...
[ 0.029154539108276367, -0.0337633453309536, -0.02038087509572506, 0.0551428347826004, 0.05057480186223984, 0.04737626761198044, 0.02709350734949112, 0.0011631966335698962, -0.039259932935237885, -0.07661300152540207, -0.025616221129894257, -0.03667965903878212, -0.043154027312994, -0.010374...
89
TextCaps: a Dataset for Image Captioning with Reading Comprehension
[ "Oleksii Sidorov", "Ronghang Hu", "Marcus Rohrbach", "Amanpreet Singh" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4358_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470732.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470732-supp.pdf
10.1007/978-3-030-58536-5_44
2003.12462
title_snapshot
Image descriptions can help visually impaired people to quickly understand the image content. While we made significant progress in automatically describing images and optical character recognition, current approaches are unable to include written text in their descriptions, although text is omnipresent in human enviro...
[ 0.007582422345876694, -0.024329064413905144, -0.017608804628252983, 0.05153656005859375, 0.050564538687467575, -0.006662608124315739, 0.0141102010384202, 0.032198816537857056, -0.0356847308576107, -0.016086839139461517, -0.06352994590997696, 0.03016963042318821, -0.052397675812244415, -0.0...
90
It is not the Journey but the Destination: Endpoint Conditioned Trajectory Prediction
[ "Karttikeya Mangalam", "Harshayu Girase", "Shreyas Agarwal", "Kuan-Hui Lee", "Ehsan Adeli", "Jitendra Malik", "Adrien Gaidon" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4423_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470749.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470749-supp.zip
10.1007/978-3-030-58536-5_45
2004.02025
title_snapshot
Human trajectory forecasting with multiple socially interact-ing agents is of critical importance for autonomous navigation in human environments, e.g., for self-driving cars and social robots. In this work, we present Predicted Endpoint Conditioned Network (PECNet) for flexible human trajectory prediction. PECNet infe...
[ -0.0023716939613223076, -0.02718701958656311, 0.016488339751958847, 0.03741986304521561, 0.028071103617548943, -0.003815937787294388, 0.029741771519184113, 0.02514035627245903, -0.007443445269018412, -0.03327222168445587, -0.03208070993423462, -0.017266957089304924, -0.06834211945533752, -...
91
Learning What to Learn for Video Object Segmentation
[ "Goutam Bhat", "Felix Järemo Lawin", "Martin Danelljan", "Andreas Robinson", "Michael Felsberg", "Luc Van Gool", "Radu Timofte" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4440_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470766.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470766-supp.zip
10.1007/978-3-030-58536-5_46
2003.11540
title_snapshot
Video object segmentation (VOS) is a highly challenging problem, since the target object is only defined by a first-frame reference mask during inference. The problem of how to capture and utilize this limited information to accurately segment the target remains a fundamental research question. We address this by intro...
[ 0.03819083049893379, 0.006021867040544748, 0.00758886244148016, 0.04142095148563385, 0.01299875508993864, 0.049151137471199036, 0.029973076656460762, 0.02044522948563099, -0.04100261256098747, -0.025206491351127625, -0.04442501440644264, 0.011286739259958267, -0.052878864109516144, 0.01175...
92
SIZER: A Dataset and Model for Parsing 3D Clothing and Learning Size Sensitive 3D Clothing
[ "Garvita Tiwari", "Bharat Lal Bhatnagar", "Tony Tung", "Gerard Pons-Moll" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4732_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480001.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480001-supp.pdf
10.1007/978-3-030-58580-8_1
2007.11610
title_snapshot
While models of 3D clothing learned from real data exist, no method can predict clothing deformation as a function of garment size. In this paper, we introduce SizerNet to predict 3D clothing conditioned on human body shape and garment size parameters, and ParserNet to infer garment meshes and shape under clothing with...
[ 0.028080936521291733, -0.020414775237441063, -0.010343859903514385, 0.00979711301624775, 0.048663314431905746, 0.04687095060944557, 0.04284847155213356, -0.0023403263185173273, -0.02856677584350109, -0.04204358160495758, -0.04866259545087814, -0.017730485647916794, -0.05959191173315048, -0...
93
LIMP: Learning Latent Shape Representations with Metric Preservation Priors
[ "Luca Cosmo", "Antonio Norelli", "Oshri Halimi", "Ron Kimmel", "Emanuele Rodolà" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4866_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480018.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480018-supp.pdf
10.1007/978-3-030-58580-8_2
2003.12283
title_snapshot
In this paper, we advocate the adoption of metric preservation as a powerful prior for learning latent representations of deformable 3D shapes. Key to our construction is the introduction of a geometric distortion criterion, defined directly on the decoded shapes, translating the preservation of the metric on the decod...
[ 0.004894479643553495, -0.02660846896469593, -0.022999389097094536, 0.0471925362944603, 0.0430520661175251, 0.05667048320174217, -0.007870599627494812, 0.011843070387840271, -0.04097440093755722, -0.07254978269338608, -0.03138921037316322, -0.017730416730046272, -0.04711677506566048, -0.002...
94
Unsupervised Sketch to Photo Synthesis
[ "Runtao Liu", "Qian Yu", "Stella X. Yu" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5277_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480035.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480035-supp.pdf
10.1007/978-3-030-58580-8_3
1909.08313
title_snapshot
Humans can envision a realistic photo given a free-hand sketch that is not only spatially imprecise and geometrically distorted but also without colors and visual details. We study unsupervised sketch to photo synthesis for the first time, learning from unpaired sketch and photo data where the target photo for a sketch...
[ 0.02882450446486473, -0.029239656403660774, -0.003909616731107235, 0.043410640209913254, 0.06767181307077408, 0.0037073444109410048, 0.011968454346060753, 0.03441965579986572, -0.02871645614504814, -0.08251681923866272, -0.04683765769004822, -0.019429439678788185, -0.07297856360673904, -0....
95
A Simple Way to Make Neural Networks Robust Against Diverse Image Corruptions
[ "Evgenia Rusak", "Lukas Schott", "Roland S. Zimmermann", "Julian Bitterwolf", "Oliver Bringmann", "Matthias Bethge", "Wieland Brendel" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5360_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480052.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480052-supp.pdf
10.1007/978-3-030-58580-8_4
2001.06057
title_snapshot
The human visual system is remarkably robust against a wide range of naturally occurring variations and corruptions like rain or snow. In contrast, the performance of modern image recognition models strongly degrades when evaluated on previously unseen corruptions. Here, we demonstrate that a simple but properly tuned ...
[ 0.02424614131450653, -0.009439312852919102, -0.021022355183959007, 0.0953536108136177, 0.03867538273334503, 0.030443590134382248, 0.011842112056910992, -0.0008240658207796514, -0.04987300932407379, -0.06179747357964516, -0.0028209423180669546, 0.00895131379365921, -0.061924561858177185, -0...
96
SoftPoolNet: Shape Descriptor for Point Cloud Completion and Classification
[ "Yida Wang", "David Joseph Tan", "Nassir Navab", "Federico Tombari" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5457_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480069.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480069-supp.pdf
10.1007/978-3-030-58580-8_5
2008.07358
title_snapshot
Point clouds are often the default choice for many applications as they exhibit more flexibility and efficiency than volumetric data. Nevertheless, their unorganized nature - points are stored in an unordered way - makes them less suited to be processed by deep learning pipelines. In this paper, we propose a method for...
[ -0.0012633624719455838, -0.03311001881957054, 0.007111954502761364, 0.034168291836977005, 0.03354945406317711, 0.05619518831372261, -0.023525793105363846, 0.04164179787039757, -0.04976264387369156, -0.06568759679794312, -0.05589841678738594, -0.021689778193831444, -0.05208534747362137, 0.0...
97
Hierarchical Face Aging through Disentangled Latent Characteristics
[ "Peipei Li", "Huaibo Huang", "Yibo Hu", "Xiang Wu", "Ran He", "Zhenan Sun" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5800_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480086.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480086-supp.pdf
10.1007/978-3-030-58580-8_6
null
null
Current age datasets lie in a long-tailed distribution, which brings difficulties to describe the aging mechanism for the imbalance ages. To alleviate it, we design a novel facial age prior to guide the aging mechanism modeling. To explore the age effects on facial images, we propose a Disentangled Adversarial Autoenco...
[ 0.009651307947933674, 0.013410812243819237, -0.005305704660713673, 0.03692132607102394, 0.03335784375667572, 0.03632538765668869, 0.03580309823155403, -0.011021273210644722, -0.0010504978708922863, -0.05224804952740669, 0.011412803083658218, 0.008414153009653091, -0.0533127523958683, 0.002...
98
Hybrid Models for Open Set Recognition
[ "Hongjie Zhang", "Ang Li", "Jie Guo", "Yanwen Guo" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5859_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480103.pdf
null
10.1007/978-3-030-58580-8_7
2003.12506
title_snapshot
Open set recognition requires a classifier to detect samples not belonging to any of the classes in its training set. Existing methods fit a probability distribution to the training samples on their embedding space and detect outliers according to this distribution. The embedding space is often obtained from a discrimi...
[ 0.01504135224968195, -0.00997193530201912, -0.017430996522307396, 0.066230408847332, 0.03254472836852074, 0.008538597263395786, 0.007771049160510302, -0.02093665674328804, -0.00922511424869299, -0.04316889867186546, -0.009455038234591484, 0.017908617854118347, -0.0999462753534317, -0.01992...
99
TopoGAN: A Topology-Aware Generative Adversarial Network
[ "Fan Wang", "Huidong Liu", "Dimitris Samaras", "Chao Chen" ]
https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5932_ECCV_2020_paper.php
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480120.pdf
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123480120-supp.pdf
10.1007/978-3-030-58580-8_8
null
null
Existing generative adversarial networks (GANs) focus on generating realistic images based on CNN-derived image features, but fail to preserve the structural properties of real images. This can be fatal in applications where the underlying structure (e.g., neurons, vessels, membranes, and road networks) of the image ca...
[ -0.005980168469250202, -0.022984329611063004, -0.0024279048666357994, 0.04034988954663277, -0.0014581397408619523, 0.016666505485773087, 0.012910713441669941, 0.02766457200050354, -0.007300356402993202, -0.059489261358976364, -0.03419167175889015, -0.03620079532265663, -0.0792614221572876, ...