paper_id
string
title
string
paper_url
string
pdf_url
string
authors
list
abstract
large_string
track
string
primary_area
string
doi
string
volume
string
issue
string
pages
string
abstract_source
string
arxiv_id
string
arxiv_id_source
string
10.1609/aaai.v35i3.16278
Dynamic to Static Lidar Scan Reconstruction Using Adversarially Trained Auto Encoder
https://ojs.aaai.org/index.php/AAAI/article/view/16278
https://ojs.aaai.org/index.php/AAAI/article/download/16278/16085
[ "Prashant Kumar", "Sabyasachi Sahoo", "Vanshil Shah", "Vineetha Kondameedi", "Abhinav Jain", "Akshaj Verma", "Chiranjib Bhattacharyya", "Vinay Vishwanath" ]
Accurate reconstruction of static environments from LiDAR scans of scenes containing dynamic objects, which we refer to as Dynamic to Static Translation (DST), is an important area of research in Autonomous Navigation. This problem has been recently explored for visual SLAM, but to the best of our knowledge no work has...
main
Computer Vision
10.1609/aaai.v35i3.16278
35
3
1836-1844
official
2105.12774
title_judge
10.1609/aaai.v35i3.16279
Regularizing Attention Networks for Anomaly Detection in Visual Question Answering
https://ojs.aaai.org/index.php/AAAI/article/view/16279
https://ojs.aaai.org/index.php/AAAI/article/download/16279/16086
[ "Doyup Lee", "Yeongjae Cheon", "Wook-Shin Han" ]
For stability and reliability of real-world applications, the robustness of DNNs in unimodal tasks has been evaluated. However, few studies consider abnormal situations that a visual question answering (VQA) model might encounter at test time after deployment in the real-world. In this study, we evaluate the robustness...
main
Computer Vision
10.1609/aaai.v35i3.16279
35
3
1845-1853
official
2009.10054
title_snapshot
10.1609/aaai.v35i3.16280
Weakly-supervised Temporal Action Localization by Uncertainty Modeling
https://ojs.aaai.org/index.php/AAAI/article/view/16280
https://ojs.aaai.org/index.php/AAAI/article/download/16280/16087
[ "Pilhyeon Lee", "Jinglu Wang", "Yan Lu", "Hyeran Byun" ]
Weakly-supervised temporal action localization aims to learn detecting temporal intervals of action classes with only video-level labels. To this end, it is crucial to separate frames of action classes from the background frames (i.e., frames not belonging to any action classes). In this paper, we present a new perspec...
main
Computer Vision
10.1609/aaai.v35i3.16280
35
3
1854-1862
official
2006.07006
title_snapshot
10.1609/aaai.v35i3.16281
Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection Consistency
https://ojs.aaai.org/index.php/AAAI/article/view/16281
https://ojs.aaai.org/index.php/AAAI/article/download/16281/16088
[ "Seokju Lee", "Sunghoon Im", "Stephen Lin", "In So Kweon" ]
We present an end-to-end joint training framework that explicitly models 6-DoF motion of multiple dynamic objects, ego-motion, and depth in a monocular camera setup without supervision. Our technical contributions are three-fold. First, we highlight the fundamental difference between inverse and forward projection whil...
main
Computer Vision
10.1609/aaai.v35i3.16281
35
3
1863-1872
official
2102.02629
title_snapshot
10.1609/aaai.v35i3.16282
Patch-Wise Attention Network for Monocular Depth Estimation
https://ojs.aaai.org/index.php/AAAI/article/view/16282
https://ojs.aaai.org/index.php/AAAI/article/download/16282/16089
[ "Sihaeng Lee", "Janghyeon Lee", "Byungju Kim", "Eojindl Yi", "Junmo Kim" ]
In computer vision, monocular depth estimation is the problem of obtaining a high-quality depth map from a two-dimensional image. This map provides information on three-dimensional scene geometry, which is necessary for various applications in academia and industry, such as robotics and autonomous driving. Recent studi...
main
Computer Vision
10.1609/aaai.v35i3.16282
35
3
1873-1881
official
null
null
10.1609/aaai.v35i4.16463
Context-Guided Adaptive Network for Efficient Human Pose Estimation
https://ojs.aaai.org/index.php/AAAI/article/view/16463
https://ojs.aaai.org/index.php/AAAI/article/download/16463/16270
[ "Lei Zhao", "Jun Wen", "Pengfei Wang", "Nenggan Zheng" ]
Although recent work has achieved great progress in human pose estimation (HPE), most methods show limitations in either inference speed or accuracy. In this paper, we propose a fast and accurate end-to-end HPE method, which is specifically designed to overcome the commonly encountered jitter box, defective box and amb...
main
Computer Vision
10.1609/aaai.v35i4.16463
35
4
3492-3499
official
null
null
10.1609/aaai.v35i4.16464
ePointDA: An End-to-End Simulation-to-Real Domain Adaptation Framework for LiDAR Point Cloud Segmentation
https://ojs.aaai.org/index.php/AAAI/article/view/16464
https://ojs.aaai.org/index.php/AAAI/article/download/16464/16271
[ "Sicheng Zhao", "Yezhen Wang", "Bo Li", "Bichen Wu", "Yang Gao", "Pengfei Xu", "Trevor Darrell", "Kurt Keutzer" ]
Due to its robust and precise distance measurements, LiDAR plays an important role in scene understanding for autonomous driving. Training deep neural networks (DNNs) on LiDAR data requires large-scale point-wise annotations, which are time-consuming and expensive to obtain. Instead, simulation-to-real domain adaptatio...
main
Computer Vision
10.1609/aaai.v35i4.16464
35
4
3500-3509
official
2009.03456
title_snapshot
10.1609/aaai.v35i4.16465
Robust Lightweight Facial Expression Recognition Network with Label Distribution Training
https://ojs.aaai.org/index.php/AAAI/article/view/16465
https://ojs.aaai.org/index.php/AAAI/article/download/16465/16272
[ "Zengqun Zhao", "Qingshan Liu", "Feng Zhou" ]
This paper presents an efficiently robust facial expression recognition (FER) network, named EfficientFace, which holds much fewer parameters but more robust to the FER in the wild. Firstly, to improve the robustness of the lightweight network, a local-feature extractor and a channel-spatial modulator are designed, in ...
main
Computer Vision
10.1609/aaai.v35i4.16465
35
4
3510-3519
official
null
null
10.1609/aaai.v35i4.16466
Joint Color-irrelevant Consistency Learning and Identity-aware Modality Adaptation for Visible-infrared Cross Modality Person Re-identification
https://ojs.aaai.org/index.php/AAAI/article/view/16466
https://ojs.aaai.org/index.php/AAAI/article/download/16466/16273
[ "Zhiwei Zhao", "Bin Liu", "Qi Chu", "Yan Lu", "Nenghai Yu" ]
Visible-infrared cross modality person re-identification (VI-ReID) is a core but challenging technology in the 24-hours intelligent surveillance system. How to eliminate the large modality gap lies in the heart of VI-ReID. Conventional methods mainly focus on directly aligning the heterogeneous modalities into the same...
main
Computer Vision
10.1609/aaai.v35i4.16466
35
4
3520-3528
official
null
null
10.1609/aaai.v35i4.16467
Robust Multi-Modality Person Re-identification
https://ojs.aaai.org/index.php/AAAI/article/view/16467
https://ojs.aaai.org/index.php/AAAI/article/download/16467/16274
[ "Aihua Zheng", "Zi Wang", "Zihan Chen", "Chenglong Li", "Jin Tang" ]
To avoid the illumination limitation in visible person re-identification (Re-ID) and the heterogeneous issue in cross-modality Re-ID, we propose to utilize complementary advantages of multiple modalities including visible (RGB), near infrared (NI) and thermal infrared (TI) ones for robust person Re-ID. A novel progress...
main
Computer Vision
10.1609/aaai.v35i4.16467
35
4
3529-3537
official
null
null
10.1609/aaai.v35i4.16468
Exploiting Sample Uncertainty for Domain Adaptive Person Re-Identification
https://ojs.aaai.org/index.php/AAAI/article/view/16468
https://ojs.aaai.org/index.php/AAAI/article/download/16468/16275
[ "Kecheng Zheng", "Cuiling Lan", "Wenjun Zeng", "Zhizheng Zhang", "Zheng-Jun Zha" ]
Many unsupervised domain adaptive (UDA) person ReID approaches combine clustering-based pseudo-label prediction with feature fine-tuning. However, because of domain gap, the pseudo-labels are not always reliable and there are noisy/incorrect labels. This would mislead the feature representation learning and deteriorate...
main
Computer Vision
10.1609/aaai.v35i4.16468
35
4
3538-3546
official
2012.08733
title_snapshot
10.1609/aaai.v35i4.16469
RESA: Recurrent Feature-Shift Aggregator for Lane Detection
https://ojs.aaai.org/index.php/AAAI/article/view/16469
https://ojs.aaai.org/index.php/AAAI/article/download/16469/16276
[ "Tu Zheng", "Hao Fang", "Yi Zhang", "Wenjian Tang", "Zheng Yang", "Haifeng Liu", "Deng Cai" ]
Lane detection is one of the most important tasks in self-driving. Due to various complex scenarios (e.g., severe occlusion, ambiguous lanes, etc.) and the sparse supervisory signals inherent in lane annotations, lane detection task is still challenging. Thus, it is difficult for the ordinary convolutional neural netwo...
main
Computer Vision
10.1609/aaai.v35i4.16469
35
4
3547-3554
official
2008.13719
title_snapshot
10.1609/aaai.v35i4.16470
CIA-SSD: Confident IoU-Aware Single-Stage Object Detector From Point Cloud
https://ojs.aaai.org/index.php/AAAI/article/view/16470
https://ojs.aaai.org/index.php/AAAI/article/download/16470/16277
[ "Wu Zheng", "Weiliang Tang", "Sijin Chen", "Li Jiang", "Chi-Wing Fu" ]
Existing single-stage detectors for locating objects in point clouds often treat object localization and category classification as separate tasks, so the localization accuracy and classification confidence may not well align. To address this issue, we present a new single-stage detector named the Confident IoU-Aware S...
main
Computer Vision
10.1609/aaai.v35i4.16470
35
4
3555-3562
official
2012.03015
title_snapshot
10.1609/aaai.v35i4.16471
Regional Attention with Architecture-Rebuilt 3D Network for RGB-D Gesture Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16471
https://ojs.aaai.org/index.php/AAAI/article/download/16471/16278
[ "Benjia Zhou", "Yunan Li", "Jun Wan" ]
Human gesture recognition has drawn much attention in the area of computer vision. However, the performance of gesture recognition is always influenced by some gesture-irrelevant factors like the background and the clothes of performers. Therefore, focusing on the regions of hand/arm is important to the gesture recogni...
main
Computer Vision
10.1609/aaai.v35i4.16471
35
4
3563-3571
official
2102.05348
title_snapshot
10.1609/aaai.v35i4.16472
Deep Semantic Dictionary Learning for Multi-label Image Classification
https://ojs.aaai.org/index.php/AAAI/article/view/16472
https://ojs.aaai.org/index.php/AAAI/article/download/16472/16279
[ "Fengtao Zhou", "Sheng Huang", "Yun Xing" ]
Compared with single-label image classification, multi-label image classification is more practical and challenging. Some recent studies attempted to leverage the semantic information of categories for improving multi-label image classification performance. However, these semantic-based methods only take semantic infor...
main
Computer Vision
10.1609/aaai.v35i4.16472
35
4
3572-3580
official
2012.12509
title_snapshot
10.1609/aaai.v35i4.16473
Model Uncertainty Guides Visual Object Tracking
https://ojs.aaai.org/index.php/AAAI/article/view/16473
https://ojs.aaai.org/index.php/AAAI/article/download/16473/16280
[ "Lijun Zhou", "Antoine Ledent", "Qintao Hu", "Ting Liu", "Jianlin Zhang", "Marius Kloft" ]
Model object trackers largely rely on the online learning of a discriminative classifier from potentially diverse sample frames. However, noisy or insufficient amounts of samples can deteriorate the classifiers' performance and cause tracking drift. Furthermore, alterations such as occlusion and blurring can cause the ...
main
Computer Vision
10.1609/aaai.v35i4.16473
35
4
3581-3589
official
null
null
10.1609/aaai.v35i4.16474
Optimizing Information Theory Based Bitwise Bottlenecks for Efficient Mixed-Precision Activation Quantization
https://ojs.aaai.org/index.php/AAAI/article/view/16474
https://ojs.aaai.org/index.php/AAAI/article/download/16474/16281
[ "Xichuan Zhou", "Kui Liu", "Cong Shi", "Haijun Liu", "Ji Liu" ]
Recent researches on information theory shed new light on the continuous attempts to open the black box of neural signal encoding. Inspired by the problem of lossy signal compression for wireless communication, this paper presents a Bitwise Bottleneck approach for quantizing and encoding neural network activations. Bas...
main
Computer Vision
10.1609/aaai.v35i4.16474
35
4
3590-3598
official
2006.05210
title_judge
10.1609/aaai.v35i4.16475
Inferring Camouflaged Objects by Texture-Aware Interactive Guidance Network
https://ojs.aaai.org/index.php/AAAI/article/view/16475
https://ojs.aaai.org/index.php/AAAI/article/download/16475/16282
[ "Jinchao Zhu", "Xiaoyu Zhang", "Shuo Zhang", "Junnan Liu" ]
Camouflaged objects, similar to the background, show indefinable boundaries and deceptive textures, which increases the difficulty of detection task and makes the model rely on features with more information. Herein, we design a texture label to facilitate our network for accurate camouflaged object segmentation. Motiv...
main
Computer Vision
10.1609/aaai.v35i4.16475
35
4
3599-3607
official
null
null
10.1609/aaai.v35i4.16476
Simple is not Easy: A Simple Strong Baseline for TextVQA and TextCaps
https://ojs.aaai.org/index.php/AAAI/article/view/16476
https://ojs.aaai.org/index.php/AAAI/article/download/16476/16283
[ "Qi Zhu", "Chenyu Gao", "Peng Wang", "Qi Wu" ]
Texts appearing in daily scenes that can be recognized by OCR (Optical Character Recognition) tools contain significant information, such as street name, product brand and prices. Two tasks -- text-based visual question answering and text-based image captioning, with a text extension from existing vision-language appli...
main
Computer Vision
10.1609/aaai.v35i4.16476
35
4
3608-3615
official
2012.05153
title_snapshot
10.1609/aaai.v35i4.16477
Fooling Thermal Infrared Pedestrian Detectors in Real World Using Small Bulbs
https://ojs.aaai.org/index.php/AAAI/article/view/16477
https://ojs.aaai.org/index.php/AAAI/article/download/16477/16284
[ "Xiaopei Zhu", "Xiao Li", "Jianmin Li", "Zheyao Wang", "Xiaolin Hu" ]
Thermal infrared detection systems play an important role in many areas such as night security, autonomous driving, and body temperature detection. They have the unique advantages of passive imaging, temperature sensitivity and penetration. But the security of these systems themselves has not been fully explored, which...
main
Computer Vision
10.1609/aaai.v35i4.16477
35
4
3616-3624
official
2101.08154
title_snapshot
10.1609/aaai.v35i4.16478
ASHF-Net: Adaptive Sampling and Hierarchical Folding Network for Robust Point Cloud Completion
https://ojs.aaai.org/index.php/AAAI/article/view/16478
https://ojs.aaai.org/index.php/AAAI/article/download/16478/16285
[ "Daoming Zong", "Shiliang Sun", "Jing Zhao" ]
Estimating the complete 3D point cloud from an incomplete one lies at the core of many vision and robotics applications. Existing methods typically predict the complete point cloud based on the global shape representation extracted from the incomplete input. Although they could predict the overall shape of 3D objects, ...
main
Computer Vision
10.1609/aaai.v35i4.16478
35
4
3625-3632
official
null
null
10.1609/aaai.v35i4.16443
Visual Tracking via Hierarchical Deep Reinforcement Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16443
https://ojs.aaai.org/index.php/AAAI/article/download/16443/16250
[ "Dawei Zhang", "Zhonglong Zheng", "Riheng Jia", "Minglu Li" ]
Visual tracking has achieved great progress due to numerous different algorithms. However, deep trackers based on classification or Siamese network still have their specific limitations. In this work, we show how to teach machines to track a generic object in videos like humans, who can use a few search steps to perfor...
main
Computer Vision
10.1609/aaai.v35i4.16443
35
4
3315-3323
official
null
null
10.1609/aaai.v35i4.16444
One for More: Selecting Generalizable Samples for Generalizable ReID Model
https://ojs.aaai.org/index.php/AAAI/article/view/16444
https://ojs.aaai.org/index.php/AAAI/article/download/16444/16251
[ "Enwei Zhang", "Xinyang Jiang", "Hao Cheng", "Ancong Wu", "Fufu Yu", "Ke Li", "Xiaowei Guo", "Feng Zheng", "Weishi Zheng", "Xing Sun" ]
Current training objectives of existing person Re-IDentification (ReID) models only ensure that the loss of the model decreases on selected training batch, with no regards to the performance on samples outside the batch. It will inevitably cause the model to over-fit the data in the dominant position (e.g., head data i...
main
Computer Vision
10.1609/aaai.v35i4.16444
35
4
3324-3332
official
2012.05475
title_snapshot
10.1609/aaai.v35i4.16445
Ada-Segment: Automated Multi-loss Adaptation for Panoptic Segmentation
https://ojs.aaai.org/index.php/AAAI/article/view/16445
https://ojs.aaai.org/index.php/AAAI/article/download/16445/16252
[ "Gengwei Zhang", "Yiming Gao", "Hang Xu", "Hao Zhang", "Zhenguo Li", "Xiaodan Liang" ]
Panoptic segmentation that unifies instance segmentation and semantic segmentation has recently attracted increasing attention. While most existing methods focus on designing novel architectures, we steer toward a different perspective: performing automated multi-loss adaptation (named Ada-Segment) on the fly to flexib...
main
Computer Vision
10.1609/aaai.v35i4.16445
35
4
3333-3341
official
2012.03603
title_snapshot
10.1609/aaai.v35i4.16446
SIMPLE: SIngle-network with Mimicking and Point Learning for Bottom-up Human Pose Estimation
https://ojs.aaai.org/index.php/AAAI/article/view/16446
https://ojs.aaai.org/index.php/AAAI/article/download/16446/16253
[ "Jiabin Zhang", "Zheng Zhu", "Jiwen Lu", "Junjie Huang", "Guan Huang", "Jie Zhou" ]
The practical application requests both accuracy and efficiency on multi-person pose estimation algorithms. But the high accuracy and fast inference speed are dominated by top-down methods and bottom-up methods respectively. To make a better trade-off between accuracy and efficiency, we propose a novel multi-person pos...
main
Computer Vision
10.1609/aaai.v35i4.16446
35
4
3342-3350
official
2104.02486
title_snapshot
10.1609/aaai.v35i4.16447
Enhancing Audio-Visual Association with Self-Supervised Curriculum Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16447
https://ojs.aaai.org/index.php/AAAI/article/download/16447/16254
[ "Jingran Zhang", "Xing Xu", "Fumin Shen", "Huimin Lu", "Xin Liu", "Heng Tao Shen" ]
The recent success of audio-visual representations learning can be largely attributed to their pervasive concurrency property, which can be used as a self-supervision signal and extract correlation information. While most recent works focus on capturing the shared associations between the audio and visual modalities, t...
main
Computer Vision
10.1609/aaai.v35i4.16447
35
4
3351-3359
official
null
null
10.1609/aaai.v35i4.16448
Unsupervised Domain Adaptation for Person Re-identification via Heterogeneous Graph Alignment
https://ojs.aaai.org/index.php/AAAI/article/view/16448
https://ojs.aaai.org/index.php/AAAI/article/download/16448/16255
[ "Minying Zhang", "Kai Liu", "Yidong Li", "Shihui Guo", "Hongtao Duan", "Yimin Long", "Yi Jin" ]
Unsupervised person re-identification (re-ID) is becoming increasingly popular due to its power in real-world systems such as public security and intelligent transportation systems. However, the person re-ID task is challenged by the problems of data distribution discrepancy across cameras and lack of label information...
main
Computer Vision
10.1609/aaai.v35i4.16448
35
4
3360-3368
official
null
null
10.1609/aaai.v35i4.16449
Proactive Privacy-preserving Learning for Retrieval
https://ojs.aaai.org/index.php/AAAI/article/view/16449
https://ojs.aaai.org/index.php/AAAI/article/download/16449/16256
[ "Peng-Fei Zhang", "Zi Huang", "Xin-Shun Xu" ]
Deep Neural Networks (DNNs) have recently achieved remarkable performance in image retrieval, yet posing great threats to data privacy. On the one hand, one may misuse a deployed DNNs based system to look up data without consent. On the other hand, organizations or individuals would legally or illegally collect data to...
main
Computer Vision
10.1609/aaai.v35i4.16449
35
4
3369-3376
official
null
null
10.1609/aaai.v35i4.16450
A Novel Visual Interpretability for Deep Neural Networks by Optimizing Activation Maps with Perturbation
https://ojs.aaai.org/index.php/AAAI/article/view/16450
https://ojs.aaai.org/index.php/AAAI/article/download/16450/16257
[ "Qinglong Zhang", "Lu Rao", "Yubin Yang" ]
Interpretability has been regarded as an essential component for deploying deep neural networks, in which the saliency-based method is one of the most prevailing interpretable approaches since it can generate individually intuitive heatmaps that highlight parts of the input image that are most important to the decision...
main
Computer Vision
10.1609/aaai.v35i4.16450
35
4
3377-3384
official
null
null
10.1609/aaai.v35i4.16451
Point Cloud Semantic Scene Completion from RGB-D Images
https://ojs.aaai.org/index.php/AAAI/article/view/16451
https://ojs.aaai.org/index.php/AAAI/article/download/16451/16258
[ "Shoulong Zhang", "Shuai Li", "Aimin Hao", "Hong Qin" ]
In this paper, we devise a novel semantic completion network, called point cloud semantic scene completion network (PCSSC-Net), for indoor scenes solely based on point clouds. Existing point cloud completion networks still suffer from their inability of fully recovering complex structures and contents from global geome...
main
Computer Vision
10.1609/aaai.v35i4.16451
35
4
3385-3393
official
null
null
10.1609/aaai.v35i4.16452
Consensus Graph Representation Learning for Better Grounded Image Captioning
https://ojs.aaai.org/index.php/AAAI/article/view/16452
https://ojs.aaai.org/index.php/AAAI/article/download/16452/16259
[ "Wenqiao Zhang", "Haochen Shi", "Siliang Tang", "Jun Xiao", "Qiang Yu", "Yueting Zhuang" ]
The contemporary visual captioning models frequently hallucinate objects that are not actually in a scene, due to the visual misclassification or over-reliance on priors that resulting in the semantic inconsistency between the visual information and the target lexical words. The most common way is to encourage the capt...
main
Computer Vision
10.1609/aaai.v35i4.16452
35
4
3394-3402
official
2112.00974
title_snapshot
10.1609/aaai.v35i4.16453
BoW Pooling: A Plug-and-Play Unit for Feature Aggregation of Point Clouds
https://ojs.aaai.org/index.php/AAAI/article/view/16453
https://ojs.aaai.org/index.php/AAAI/article/download/16453/16260
[ "Xiang Zhang", "Xiao Sun", "Zhouhui Lian" ]
Point cloud provides a compact and flexible representation for 3D shapes and recently attracts more and more attention due to the increasing demands in practical applications. The major challenge of handling such irregular data is how to achieve the permutation invariance of points in the input. Most of existing method...
main
Computer Vision
10.1609/aaai.v35i4.16453
35
4
3403-3411
official
null
null
10.1609/aaai.v35i4.16454
Diverse Knowledge Distillation for End-to-End Person Search
https://ojs.aaai.org/index.php/AAAI/article/view/16454
https://ojs.aaai.org/index.php/AAAI/article/download/16454/16261
[ "Xinyu Zhang", "Xinlong Wang", "Jia-Wang Bian", "Chunhua Shen", "Mingyu You" ]
Person search aims to localize and identify a specific person from a gallery of images. Recent methods can be categorized into two groups, i.e., two-step and end-to-end approaches. The former views person search as two independent tasks and achieves dominant results using separately trained person detection and re-iden...
main
Computer Vision
10.1609/aaai.v35i4.16454
35
4
3412-3420
official
2012.11187
title_snapshot
10.1609/aaai.v35i4.16455
Weakly Supervised Semantic Segmentation for Large-Scale Point Cloud
https://ojs.aaai.org/index.php/AAAI/article/view/16455
https://ojs.aaai.org/index.php/AAAI/article/download/16455/16262
[ "Yachao Zhang", "Zonghao Li", "Yuan Xie", "Yanyun Qu", "Cuihua Li", "Tao Mei" ]
Existing methods for large-scale point cloud semantic segmentation require expensive, tedious and error-prone manual point-wise annotation. Intuitively, weakly supervised training is a direct solution to reduce the labeling costs. However, for weakly supervised large-scale point cloud semantic segmentation, too few ann...
main
Computer Vision
10.1609/aaai.v35i4.16455
35
4
3421-3429
official
2212.04744
title_snapshot
10.1609/aaai.v35i4.16456
PC-RGNN: Point Cloud Completion and Graph Neural Network for 3D Object Detection
https://ojs.aaai.org/index.php/AAAI/article/view/16456
https://ojs.aaai.org/index.php/AAAI/article/download/16456/16263
[ "Yanan Zhang", "Di Huang", "Yunhong Wang" ]
LiDAR-based 3D object detection is an important task for autonomous driving and current approaches suffer from sparse and partial point clouds caused by distant and occluded objects. In this paper, we propose a novel two-stage framework, namely PC-RGNN, which deals with these challenges by two specific solutions. On th...
main
Computer Vision
10.1609/aaai.v35i4.16456
35
4
3430-3437
official
2012.10412
title_snapshot
10.1609/aaai.v35i4.16457
Efficient License Plate Recognition via Holistic Position Attention
https://ojs.aaai.org/index.php/AAAI/article/view/16457
https://ojs.aaai.org/index.php/AAAI/article/download/16457/16264
[ "Yesheng Zhang", "Zilei Wang", "Jiafan Zhuang" ]
License plate recognition (LPR) is a fundamental component of various intelligent transportation systems, and is always expected to be accurate and efficient enough in real-world applications. Nowadays, recognition of single character has been sophisticated benefiting from the power of deep learning, and extracting pos...
main
Computer Vision
10.1609/aaai.v35i4.16457
35
4
3438-3446
official
null
null
10.1609/aaai.v35i4.16458
Bag of Tricks for Long-Tailed Visual Recognition with Deep Convolutional Neural Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16458
https://ojs.aaai.org/index.php/AAAI/article/download/16458/16265
[ "Yongshun Zhang", "Xiu-Shen Wei", "Boyan Zhou", "Jianxin Wu" ]
In recent years, visual recognition on challenging long-tailed distributions, where classes often exhibit extremely imbalanced frequencies, has made great progress mostly based on various complex paradigms (e.g., meta learning). Apart from these complex methods, simple refinements on training procedures also make contr...
main
Computer Vision
10.1609/aaai.v35i4.16458
35
4
3447-3455
official
null
null
10.1609/aaai.v35i4.16459
Depth Privileged Object Detection in Indoor Scenes via Deformation Hallucination
https://ojs.aaai.org/index.php/AAAI/article/view/16459
https://ojs.aaai.org/index.php/AAAI/article/download/16459/16266
[ "Zhijie Zhang", "Yan Liu", "Junjie Chen", "Li Niu", "Liqing Zhang" ]
RGB-D object detection has achieved significant advance, because depth provides complementary geometric information to RGB images. Considering depth images are unavailable in some scenarios, we focus on depth privileged object detection in indoor scenes, where the depth images are only available in the training phase. ...
main
Computer Vision
10.1609/aaai.v35i4.16459
35
4
3456-3464
official
null
null
10.1609/aaai.v35i4.16460
Learning Flexibly Distributional Representation for Low-quality 3D Face Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16460
https://ojs.aaai.org/index.php/AAAI/article/download/16460/16267
[ "Zihui Zhang", "Cuican Yu", "Shuang Xu", "Huibin Li" ]
Due to the superiority of using geometric information, 3D Face Recognition (FR) has achieved great successes. Existing methods focus on high-quality 3D FR which is unpractical in real scenarios. Low-quality 3D FR is a more realistic scenario but the low-quality data are born with heavy noises. Therefore, exploring nois...
main
Computer Vision
10.1609/aaai.v35i4.16460
35
4
3465-3473
official
null
null
10.1609/aaai.v35i4.16461
IA-GM: A Deep Bidirectional Learning Method for Graph Matching
https://ojs.aaai.org/index.php/AAAI/article/view/16461
https://ojs.aaai.org/index.php/AAAI/article/download/16461/16268
[ "Kaixuan Zhao", "Shikui Tu", "Lei Xu" ]
Existing deep learning methods for graph matching(GM) problems usually considered affinity learningto assist combinatorial optimization in a feedforward pipeline, and parameter learning is executed by back-propagating the gradients of the matching loss. Such a pipeline pays little attention to the possible complementar...
main
Computer Vision
10.1609/aaai.v35i4.16461
35
4
3474-3482
official
null
null
10.1609/aaai.v35i4.16462
Distribution Adaptive INT8 Quantization for Training CNNs
https://ojs.aaai.org/index.php/AAAI/article/view/16462
https://ojs.aaai.org/index.php/AAAI/article/download/16462/16269
[ "Kang Zhao", "Sida Huang", "Pan Pan", "Yinghan Li", "Yingya Zhang", "Zhenyu Gu", "Yinghui Xu" ]
Researches have demonstrated that low bit-width (e.g., INT8) quantization can be employed to accelerate the inference process. It makes the gradient quantization very promising since the backward propagation requires approximately twice more computation than forward one. Due to the variability and uncertainty of gradie...
main
Computer Vision
10.1609/aaai.v35i4.16462
35
4
3483-3491
official
2102.04782
title_snapshot
10.1609/aaai.v35i4.16423
Object Relation Attention for Image Paragraph Captioning
https://ojs.aaai.org/index.php/AAAI/article/view/16423
https://ojs.aaai.org/index.php/AAAI/article/download/16423/16230
[ "Li-Chuan Yang", "Chih-Yuan Yang", "Jane Yung-jen Hsu" ]
Image paragraph captioning aims to automatically generate a paragraph from a given image. It is an extension of image captioning in terms of generating multiple sentences instead of a single one, and it is more challenging because paragraphs are longer, more informative, and more linguistically complicated. Because a p...
main
Computer Vision
10.1609/aaai.v35i4.16423
35
4
3136-3144
official
null
null
10.1609/aaai.v35i4.16424
Adversarial Robustness through Disentangled Representations
https://ojs.aaai.org/index.php/AAAI/article/view/16424
https://ojs.aaai.org/index.php/AAAI/article/download/16424/16231
[ "Shuo Yang", "Tianyu Guo", "Yunhe Wang", "Chang Xu" ]
Despite the remarkable empirical performance of deep learning models, their vulnerability to adversarial examples has been revealed in many studies. They are prone to make a susceptible prediction to the input with imperceptible adversarial perturbation. Although recent works have remarkably improved the model's robust...
main
Computer Vision
10.1609/aaai.v35i4.16424
35
4
3145-3153
official
null
null
10.1609/aaai.v35i4.16425
CPCGAN: A Controllable 3D Point Cloud Generative Adversarial Network with Semantic Label Generating
https://ojs.aaai.org/index.php/AAAI/article/view/16425
https://ojs.aaai.org/index.php/AAAI/article/download/16425/16232
[ "Ximing Yang", "Yuan Wu", "Kaiyi Zhang", "Cheng Jin" ]
Generative Adversarial Networks (GAN) are good at generating variant samples of complex data distributions. Generating a sample with certain properties is one of the major tasks in the real-world application of GANs. In this paper, we propose a novel generative adversarial network to generate 3D point clouds from rando...
main
Computer Vision
10.1609/aaai.v35i4.16425
35
4
3154-3162
official
null
null
10.1609/aaai.v35i4.16426
R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object
https://ojs.aaai.org/index.php/AAAI/article/view/16426
https://ojs.aaai.org/index.php/AAAI/article/download/16426/16233
[ "Xue Yang", "Junchi Yan", "Ziming Feng", "Tao He" ]
Rotation detection is a challenging task due to the difficulties of locating the multi-angle objects and separating them effectively from the background. Though considerable progress has been made, for practical settings, there still exist challenges for rotating objects with large aspect ratio, dense distribution and ...
main
Computer Vision
10.1609/aaai.v35i4.16426
35
4
3163-3171
official
1908.05612
title_snapshot
10.1609/aaai.v35i4.16427
One-shot Face Reenactment Using Appearance Adaptive Normalization
https://ojs.aaai.org/index.php/AAAI/article/view/16427
https://ojs.aaai.org/index.php/AAAI/article/download/16427/16234
[ "Guangming Yao", "Yi Yuan", "Tianjia Shao", "Shuang Li", "Shanqi Liu", "Yong Liu", "Mengmeng Wang", "Kun Zhou" ]
The paper proposes a novel generative adversarial network for one-shot face reenactment, which can animate a single face image to a different pose-and-expression (provided by a driving image) while keeping its original appearance. The core of our network is a novel mechanism called appearance adaptive normalization, wh...
main
Computer Vision
10.1609/aaai.v35i4.16427
35
4
3172-3180
official
2102.03984
title_snapshot
10.1609/aaai.v35i4.16428
A Case Study of the Shortcut Effects in Visual Commonsense Reasoning
https://ojs.aaai.org/index.php/AAAI/article/view/16428
https://ojs.aaai.org/index.php/AAAI/article/download/16428/16235
[ "Keren Ye", "Adriana Kovashka" ]
Visual reasoning and question-answering have gathered attention in recent years. Many datasets and evaluation protocols have been proposed; some have been shown to contain bias that allows models to ``cheat'' without performing true, generalizable reasoning. A well-known bias is dependence on language priors (frequency...
main
Computer Vision
10.1609/aaai.v35i4.16428
35
4
3181-3189
official
null
null
10.1609/aaai.v35i4.16429
Instance Mining with Class Feature Banks for Weakly Supervised Object Detection
https://ojs.aaai.org/index.php/AAAI/article/view/16429
https://ojs.aaai.org/index.php/AAAI/article/download/16429/16236
[ "Yufei Yin", "Jiajun Deng", "Wengang Zhou", "Houqiang Li" ]
Recent progress on weakly supervised object detection (WSOD) is characterized by formulating WSOD as a Multiple Instance Learning (MIL) problem and taking online refinement with the selected region proposals from MIL. However, MIL inclines to select the most discriminative part rather than the entire instance as the to...
main
Computer Vision
10.1609/aaai.v35i4.16429
35
4
3190-3198
official
null
null
10.1609/aaai.v35i4.16430
Multimodal Fusion via Teacher-Student Network for Indoor Action Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16430
https://ojs.aaai.org/index.php/AAAI/article/download/16430/16237
[ "Bruce X.B. Yu", "Yan Liu", "Keith C.C. Chan" ]
Indoor action recognition plays an important role in modern society, such as intelligent healthcare in large mobile cabin hospitals. With the wide usage of depth sensors like Kinect, multimodal information including skeleton and RGB modalities brings a promising way to improve the performance. However, existing methods...
main
Computer Vision
10.1609/aaai.v35i4.16430
35
4
3199-3207
official
null
null
10.1609/aaai.v35i4.16431
ERNIE-ViL: Knowledge Enhanced Vision-Language Representations through Scene Graphs
https://ojs.aaai.org/index.php/AAAI/article/view/16431
https://ojs.aaai.org/index.php/AAAI/article/download/16431/16238
[ "Fei Yu", "Jiji Tang", "Weichong Yin", "Yu Sun", "Hao Tian", "Hua Wu", "Haifeng Wang" ]
We propose a knowledge-enhanced approach, ERNIE-ViL, which incorporates structured knowledge obtained from scene graphs to learn joint representations of vision-language. ERNIE-ViL tries to build the detailed semantic connections (objects, attributes of objects and relationships between objects) across vision and langu...
main
Computer Vision
10.1609/aaai.v35i4.16431
35
4
3208-3216
official
2006.16934
title_judge
10.1609/aaai.v35i4.16432
High-Resolution Deep Image Matting
https://ojs.aaai.org/index.php/AAAI/article/view/16432
https://ojs.aaai.org/index.php/AAAI/article/download/16432/16239
[ "Haichao Yu", "Ning Xu", "Zilong Huang", "Yuqian Zhou", "Humphrey Shi" ]
Image matting is a key technique for image and video editing and composition. Conventionally, deep learning approaches take the whole input image and an associated trimap to infer the alpha matte using convolutional neural networks. Such approaches set state-of-the-arts in image matting; however, they may fail in real-...
main
Computer Vision
10.1609/aaai.v35i4.16432
35
4
3217-3224
official
2009.06613
title_snapshot
10.1609/aaai.v35i4.16433
CAKES: Channel-wise Automatic KErnel Shrinking for Efficient 3D Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16433
https://ojs.aaai.org/index.php/AAAI/article/download/16433/16240
[ "Qihang Yu", "Yingwei Li", "Jieru Mei", "Yuyin Zhou", "Alan Yuille" ]
3D Convolution Neural Networks (CNNs) have been widely applied to 3D scene understanding, such as video analysis and volumetric image recognition. However, 3D networks can easily lead to over-parameterization which incurs expensive computation cost. In this paper, we propose Channel-wise Automatic KErnel Shrinking (CAK...
main
Computer Vision
10.1609/aaai.v35i4.16433
35
4
3225-3233
official
2003.12798
title_snapshot
10.1609/aaai.v35i4.16434
Structure-Consistent Weakly Supervised Salient Object Detection with Local Saliency Coherence
https://ojs.aaai.org/index.php/AAAI/article/view/16434
https://ojs.aaai.org/index.php/AAAI/article/download/16434/16241
[ "Siyue Yu", "Bingfeng Zhang", "Jimin Xiao", "Eng Gee Lim" ]
Sparse labels have been attracting much attention in recent years. However, the performance gap between weakly supervised and fully supervised salient object detection methods is huge, and most previous weakly supervised works adopt complex training methods with many bells and whistles. In this work, we propose a one-r...
main
Computer Vision
10.1609/aaai.v35i4.16434
35
4
3234-3242
official
2012.04404
title_snapshot
10.1609/aaai.v35i4.16435
Fast and Compact Bilinear Pooling by Shifted Random Maclaurin
https://ojs.aaai.org/index.php/AAAI/article/view/16435
https://ojs.aaai.org/index.php/AAAI/article/download/16435/16242
[ "Tan Yu", "Xiaoyun Li", "Ping Li" ]
Bilinear pooling has achieved an excellent performance in many computer vision tasks such as fine-grained classification, scene recognition and texture recognition. However, the high-dimension features from bilinear pooling can sometimes be inefficient and prone to over-fitting. Random Maclaurin (RM) is a widely used G...
main
Computer Vision
10.1609/aaai.v35i4.16435
35
4
3243-3251
official
null
null
10.1609/aaai.v35i4.16436
Simple and Effective Stochastic Neural Networks
https://ojs.aaai.org/index.php/AAAI/article/view/16436
https://ojs.aaai.org/index.php/AAAI/article/download/16436/16243
[ "Tianyuan Yu", "Yongxin Yang", "Da Li", "Timothy Hospedales", "Tao Xiang" ]
Stochastic neural networks (SNNs) are currently topical, with several paradigms being actively investigated including dropout, Bayesian neural networks, variational information bottleneck (VIB) and noise regularized learning. These neural network variants impact several major considerations, including generalization, n...
main
Computer Vision
10.1609/aaai.v35i4.16436
35
4
3252-3260
official
null
null
10.1609/aaai.v35i4.16437
Learning Visual Context for Group Activity Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16437
https://ojs.aaai.org/index.php/AAAI/article/download/16437/16244
[ "Hangjie Yuan", "Dong Ni" ]
Group activity recognition aims to recognize an overall activity in a multi-person scene. Previous methods strive to reason on individual features. However, they under-explore the person-specific contextual information, which is significant and informative in computer vision tasks. In this paper, we propose a new reaso...
main
Computer Vision
10.1609/aaai.v35i4.16437
35
4
3261-3269
official
null
null
10.1609/aaai.v35i4.16438
StrokeGAN: Reducing Mode Collapse in Chinese Font Generation via Stroke Encoding
https://ojs.aaai.org/index.php/AAAI/article/view/16438
https://ojs.aaai.org/index.php/AAAI/article/download/16438/16245
[ "Jinshan Zeng", "Qi Chen", "Yunxin Liu", "Mingwen Wang", "Yuan Yao" ]
The generation of stylish Chinese fonts is an important problem involved in many applications. Most of existing generation methods are based on the deep generative models, particularly, the generative adversarial networks (GAN) based models. However, these deep generative models may suffer from the mode collapse issue,...
main
Computer Vision
10.1609/aaai.v35i4.16438
35
4
3270-3277
official
2012.08687
title_snapshot
10.1609/aaai.v35i4.16439
Demodalizing Face Recognition with Synthetic Samples
https://ojs.aaai.org/index.php/AAAI/article/view/16439
https://ojs.aaai.org/index.php/AAAI/article/download/16439/16246
[ "Zhonghua Zhai", "Pengju Yang", "Xiaofeng Zhang", "Maji Huang", "Haijing Cheng", "Xuejun Yan", "Chunmao Wang", "Shiliang Pu" ]
Using data generated by generative adversarial networks or three-dimensional (3D) technology for face recognition training is a theoretically reasonable solution to the problems of unbalanced data distributions and data scarcity. However, due to the modal difference between synthetic data and real data, the direct use ...
main
Computer Vision
10.1609/aaai.v35i4.16439
35
4
3278-3286
official
null
null
10.1609/aaai.v35i4.16440
EMLight: Lighting Estimation via Spherical Distribution Approximation
https://ojs.aaai.org/index.php/AAAI/article/view/16440
https://ojs.aaai.org/index.php/AAAI/article/download/16440/16247
[ "Fangneng Zhan", "Changgong Zhang", "Yingchen Yu", "Yuan Chang", "Shijian Lu", "Feiying Ma", "Xuansong Xie" ]
Illumination estimation from a single image is critical in 3D rendering and it has been investigated extensively in the computer vision and computer graphic research community. On the other hand, existing works estimate illumination by either regressing light parameters or generating illumination maps that are often ha...
main
Computer Vision
10.1609/aaai.v35i4.16440
35
4
3287-3295
official
2012.11116
title_snapshot
10.1609/aaai.v35i4.16441
Universal Adversarial Perturbations Through the Lens of Deep Steganography: Towards a Fourier Perspective
https://ojs.aaai.org/index.php/AAAI/article/view/16441
https://ojs.aaai.org/index.php/AAAI/article/download/16441/16248
[ "Chaoning Zhang", "Philipp Benz", "Adil Karjauv", "In So Kweon" ]
The booming interest in adversarial attacks stems from a misalignment between human vision and a deep neural network (DNN), \ie~a human imperceptible perturbation fools the DNN. Moreover, a single perturbation, often called universal adversarial perturbation (UAP), can be generated to fool the DNN for most images. A si...
main
Computer Vision
10.1609/aaai.v35i4.16441
35
4
3296-3304
official
2102.06479
title_snapshot
10.1609/aaai.v35i4.16442
SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16442
https://ojs.aaai.org/index.php/AAAI/article/download/16442/16249
[ "Chengwei Zhang", "Yunlu Xu", "Zhanzhan Cheng", "Shiliang Pu", "Yi Niu", "Fei Wu", "Futai Zou" ]
Arbitrary text appearance poses a great challenge in scene text recognition tasks. Existing works mostly handle with the problem in consideration of the shape distortion, including perspective distortions, line curvature or other style variations. Rectification (i.e., spatial transformers) as the preprocessing stage is...
main
Computer Vision
10.1609/aaai.v35i4.16442
35
4
3305-3314
official
2005.13117
title_snapshot
10.1609/aaai.v35i4.16403
Binaural Audio-Visual Localization
https://ojs.aaai.org/index.php/AAAI/article/view/16403
https://ojs.aaai.org/index.php/AAAI/article/download/16403/16210
[ "Xinyi Wu", "Zhenyao Wu", "Lili Ju", "Song Wang" ]
Localizing sound sources in a visual scene has many important applications and quite a few traditional or learning-based methods have been proposed for this task. Humans have the ability to roughly localize sound sources within or beyond the range of the vision using their binaural system. However most existing methods...
main
Computer Vision
10.1609/aaai.v35i4.16403
35
4
2961-2968
official
null
null
10.1609/aaai.v35i4.16404
Beating Attackers At Their Own Games: Adversarial Example Detection Using Adversarial Gradient Directions
https://ojs.aaai.org/index.php/AAAI/article/view/16404
https://ojs.aaai.org/index.php/AAAI/article/download/16404/16211
[ "Yuhang Wu", "Sunpreet S Arora", "Yanhong Wu", "Hao Yang" ]
Adversarial examples are input examples that are specifically crafted to deceive machine learning classifiers. State-of-the-art adversarial example detection methods characterize an input example as adversarial either by quantifying the magnitude of feature variations under multiple perturbations or by measuring its di...
main
Computer Vision
10.1609/aaai.v35i4.16404
35
4
2969-2977
official
2012.15386
title_snapshot
10.1609/aaai.v35i4.16405
Shape-Pose Ambiguity in Learning 3D Reconstruction from Images
https://ojs.aaai.org/index.php/AAAI/article/view/16405
https://ojs.aaai.org/index.php/AAAI/article/download/16405/16212
[ "Yunjie Wu", "Zhengxing Sun", "Youcheng Song", "Yunhan Sun", "YiJie Zhong", "Jinlong Shi" ]
Learning single-image 3D reconstruction with only 2D images supervision is a promising research topic. The main challenge in image-supervised 3D reconstruction is the shape-pose ambiguity, which means a 2D supervision can be explained by an erroneous 3D shape from an erroneous pose. It will introduce high uncertainty a...
main
Computer Vision
10.1609/aaai.v35i4.16405
35
4
2978-2985
official
null
null
10.1609/aaai.v35i4.16406
Boundary Proposal Network for Two-stage Natural Language Video Localization
https://ojs.aaai.org/index.php/AAAI/article/view/16406
https://ojs.aaai.org/index.php/AAAI/article/download/16406/16213
[ "Shaoning Xiao", "Long Chen", "Songyang Zhang", "Wei Ji", "Jian Shao", "Lu Ye", "Jun Xiao" ]
We aim to address the problem of Natural Language Video Localization (NLVL) — localizing the video segment corresponding to a natural language description in a long and untrimmed video. State-of-the-art NLVL methods are almost in one-stage fashion, which can be typically grouped into two categories: 1) anchor-based app...
main
Computer Vision
10.1609/aaai.v35i4.16406
35
4
2986-2994
official
2103.08109
title_snapshot
10.1609/aaai.v35i4.16407
Amodal Segmentation Based on Visible Region Segmentation and Shape Prior
https://ojs.aaai.org/index.php/AAAI/article/view/16407
https://ojs.aaai.org/index.php/AAAI/article/download/16407/16214
[ "Yuting Xiao", "Yanyu Xu", "Ziming Zhong", "Weixin Luo", "Jiawei Li", "Shenghua Gao" ]
Almost all existing amodal segmentation methods make the inferences of occluded regions by using features corresponding to the whole image. This is against the human's amodal perception, where human uses the visible part and the shape prior knowledge of the target to infer the occluded region. To mimic the behavior of ...
main
Computer Vision
10.1609/aaai.v35i4.16407
35
4
2995-3003
official
2012.05598
title_snapshot
10.1609/aaai.v35i4.16408
Locate Globally, Segment Locally: A Progressive Architecture With Knowledge Review Network for Salient Object Detection
https://ojs.aaai.org/index.php/AAAI/article/view/16408
https://ojs.aaai.org/index.php/AAAI/article/download/16408/16215
[ "Binwei Xu", "Haoran Liang", "Ronghua Liang", "Peng Chen" ]
Salient object location and segmentation are two different tasks in salient object detection (SOD). The former aims to globally find the most attractive objects in an image, whereas the latter can be achieved only using local regions that contain salient objects. However, previous methods mainly accomplish the two task...
main
Computer Vision
10.1609/aaai.v35i4.16408
35
4
3004-3012
official
null
null
10.1609/aaai.v35i4.16409
Invariant Teacher and Equivariant Student for Unsupervised 3D Human Pose Estimation
https://ojs.aaai.org/index.php/AAAI/article/view/16409
https://ojs.aaai.org/index.php/AAAI/article/download/16409/16216
[ "Chenxin Xu", "Siheng Chen", "Maosen Li", "Ya Zhang" ]
We propose a novel method based on teacher-student learning framework for 3D human pose estimation without any 3D annotation or side information. To solve this unsupervised-learning problem, the teacher network adopts pose-dictionary-based modeling for regularization to estimate a physically plausible 3D pose. To handl...
main
Computer Vision
10.1609/aaai.v35i4.16409
35
4
3013-3021
official
2012.09398
title_snapshot
10.1609/aaai.v35i4.16410
Imagine, Reason and Write: Visual Storytelling with Graph Knowledge and Relational Reasoning
https://ojs.aaai.org/index.php/AAAI/article/view/16410
https://ojs.aaai.org/index.php/AAAI/article/download/16410/16217
[ "Chunpu Xu", "Min Yang", "Chengming Li", "Ying Shen", "Xiang Ao", "Ruifeng Xu" ]
Visual storytelling is a task of creating a short story based on photo streams. Different from visual captions, stories contain not only factual descriptions, but also imaginary concepts that do not appear in the images. In this paper, we propose a novel imagine-reason-write generation framework (IRW) for visual storyt...
main
Computer Vision
10.1609/aaai.v35i4.16410
35
4
3022-3029
official
null
null
10.1609/aaai.v35i4.16411
Self-supervised Multi-view Stereo via Effective Co-Segmentation and Data-Augmentation
https://ojs.aaai.org/index.php/AAAI/article/view/16411
https://ojs.aaai.org/index.php/AAAI/article/download/16411/16218
[ "Hongbin Xu", "Zhipeng Zhou", "Yu Qiao", "Wenxiong Kang", "Qiuxia Wu" ]
Recent studies have witnessed that self-supervised methods based on view synthesis obtain clear progress on multi-view stereo (MVS). However, existing methods rely on the assumption that the corresponding points among different views share the same color, which may not always be true in practice. This may lead to unrel...
main
Computer Vision
10.1609/aaai.v35i4.16411
35
4
3030-3038
official
2104.05374
title_snapshot
10.1609/aaai.v35i4.16412
Efficient Deep Image Denoising via Class Specific Convolution
https://ojs.aaai.org/index.php/AAAI/article/view/16412
https://ojs.aaai.org/index.php/AAAI/article/download/16412/16219
[ "Lu Xu", "Jiawei Zhang", "Xuanye Cheng", "Feng Zhang", "Xing Wei", "Jimmy Ren" ]
Deep neural networks have been widely used in image denoising during the past few years. Even though they achieve great success on this problem, they are computationally inefficient which makes them inappropriate to be implemented in mobile devices. In this paper, we propose an efficient deep neural network for image d...
main
Computer Vision
10.1609/aaai.v35i4.16412
35
4
3039-3046
official
2103.01624
title_snapshot
10.1609/aaai.v35i4.16413
Investigate Indistinguishable Points in Semantic Segmentation of 3D Point Cloud
https://ojs.aaai.org/index.php/AAAI/article/view/16413
https://ojs.aaai.org/index.php/AAAI/article/download/16413/16220
[ "Mingye Xu", "Zhipeng Zhou", "Junhao Zhang", "Yu Qiao" ]
This paper investigates the indistinguishable points (difficult to predict label) in semantic segmentation for large-scale 3D point clouds. The indistinguishable points consist of those located in complex boundary, points with similar local textures but different categories, and points in isolate small hard areas, whic...
main
Computer Vision
10.1609/aaai.v35i4.16413
35
4
3047-3055
official
2103.10339
title_snapshot
10.1609/aaai.v35i4.16414
Learning Geometry-Disentangled Representation for Complementary Understanding of 3D Object Point Cloud
https://ojs.aaai.org/index.php/AAAI/article/view/16414
https://ojs.aaai.org/index.php/AAAI/article/download/16414/16221
[ "Mutian Xu", "Junhao Zhang", "Zhipeng Zhou", "Mingye Xu", "Xiaojuan Qi", "Yu Qiao" ]
In 2D image processing, some attempts decompose images into high and low frequency components for describing edge and smooth parts respectively. Similarly, the contour and flat area of 3D objects, such as the boundary and seat area of a chair, describe different but also complementary geometries. However, such investig...
main
Computer Vision
10.1609/aaai.v35i4.16414
35
4
3056-3064
official
2012.10921
title_snapshot
10.1609/aaai.v35i4.16415
Searching for Alignment in Face Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16415
https://ojs.aaai.org/index.php/AAAI/article/download/16415/16222
[ "Xiaqing Xu", "Qiang Meng", "Yunxiao Qin", "Jianzhu Guo", "Chenxu Zhao", "Feng Zhou", "Zhen Lei" ]
A standard pipeline of current face recognition frameworks consists of four individual steps: locating a face with a rough bounding box and several fiducial landmarks, aligning the face image using a pre-defined template, extracting representations and comparing. Among them, face detection, landmark detection and repre...
main
Computer Vision
10.1609/aaai.v35i4.16415
35
4
3065-3073
official
2102.05447
title_snapshot
10.1609/aaai.v35i4.16416
GIF Thumbnails: Attract More Clicks to Your Videos
https://ojs.aaai.org/index.php/AAAI/article/view/16416
https://ojs.aaai.org/index.php/AAAI/article/download/16416/16223
[ "Yi Xu", "Fan Bai", "Yingxuan Shi", "Qiuyu Chen", "Longwen Gao", "Kai Tian", "Shuigeng Zhou", "Huyang Sun" ]
With the rapid increase of mobile devices and online media, more and more people prefer posting/viewing videos online. Generally, these videos are presented on video streaming sites with image thumbnails and text titles. While facing huge amounts of videos, a viewer clicks through a certain video with high probability ...
main
Computer Vision
10.1609/aaai.v35i4.16416
35
4
3074-3082
official
null
null
10.1609/aaai.v35i4.16417
FaceController: Controllable Attribute Editing for Face in the Wild
https://ojs.aaai.org/index.php/AAAI/article/view/16417
https://ojs.aaai.org/index.php/AAAI/article/download/16417/16224
[ "Zhiliang Xu", "Xiyu Yu", "Zhibin Hong", "Zhen Zhu", "Junyu Han", "Jingtuo Liu", "Errui Ding", "Xiang Bai" ]
Face attribute editing aims to generate faces with one or multiple desired face attributes manipulated while other details are preserved. Unlike prior works such as GAN inversion which has an expensive reverse mapping process, we propose a simple feed-forward network to generate high-fidelity manipulated faces. By simp...
main
Computer Vision
10.1609/aaai.v35i4.16417
35
4
3083-3091
official
2102.11464
title_snapshot
10.1609/aaai.v35i4.16418
AnchorFace: An Anchor-based Facial Landmark Detector Across Large Poses
https://ojs.aaai.org/index.php/AAAI/article/view/16418
https://ojs.aaai.org/index.php/AAAI/article/download/16418/16225
[ "Zixuan Xu", "Banghuai Li", "Ye Yuan", "Miao Geng" ]
Facial landmark localization aims to detect the predefined points of human faces, and the topic has been rapidly improved with the recent development of neural network based methods. However, it remains a challenging task when dealing with faces in unconstrained scenarios, especially with large pose variations. In this...
main
Computer Vision
10.1609/aaai.v35i4.16418
35
4
3092-3100
official
2007.03221
title_snapshot
10.1609/aaai.v35i4.16419
Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion
https://ojs.aaai.org/index.php/AAAI/article/view/16419
https://ojs.aaai.org/index.php/AAAI/article/download/16419/16226
[ "Xu Yan", "Jiantao Gao", "Jie Li", "Ruimao Zhang", "Zhen Li", "Rui Huang", "Shuguang Cui" ]
LiDAR point cloud analysis is a core task for 3D computer vision, especially for autonomous driving. However, due to the severe sparsity and noise interference in the single sweep LiDAR point cloud, the accurate semantic segmentation is non-trivial to achieve. In this paper, we propose a novel sparse LiDAR point cloud ...
main
Computer Vision
10.1609/aaai.v35i4.16419
35
4
3101-3109
official
2012.03762
title_snapshot
10.1609/aaai.v35i4.16420
Learning Semantic Context from Normal Samples for Unsupervised Anomaly Detection
https://ojs.aaai.org/index.php/AAAI/article/view/16420
https://ojs.aaai.org/index.php/AAAI/article/download/16420/16227
[ "Xudong Yan", "Huaidong Zhang", "Xuemiao Xu", "Xiaowei Hu", "Pheng-Ann Heng" ]
Unsupervised anomaly detection aims to identify data samples that have low probability density from a set of input samples, and only the normal samples are provided for model training. The inference of abnormal regions on the input image requires an understanding of the surrounding semantic context. This work presents ...
main
Computer Vision
10.1609/aaai.v35i4.16420
35
4
3110-3118
official
null
null
10.1609/aaai.v35i4.16421
Non-Autoregressive Coarse-to-Fine Video Captioning
https://ojs.aaai.org/index.php/AAAI/article/view/16421
https://ojs.aaai.org/index.php/AAAI/article/download/16421/16228
[ "Bang Yang", "Yuexian Zou", "Fenglin Liu", "Can Zhang" ]
It is encouraged to see that progress has been made to bridge videos and natural language. However, mainstream video captioning methods suffer from slow inference speed due to the sequential manner of autoregressive decoding, and prefer generating generic descriptions due to the insufficient training of visual words (e...
main
Computer Vision
10.1609/aaai.v35i4.16421
35
4
3119-3127
official
1911.12018
title_snapshot
10.1609/aaai.v35i4.16422
Learning to Attack Real-World Models for Person Re-identification via Virtual-Guided Meta-Learning
https://ojs.aaai.org/index.php/AAAI/article/view/16422
https://ojs.aaai.org/index.php/AAAI/article/download/16422/16229
[ "Fengxiang Yang", "Zhun Zhong", "Hong Liu", "Zheng Wang", "Zhiming Luo", "Shaozi Li", "Nicu Sebe", "Shin'ichi Satoh" ]
Recent advances in person re-identification (re-ID) have led to impressive retrieval accuracy. However, existing re-ID models are challenged by the adversarial examples crafted by adding quasi-imperceptible perturbations. Moreover, re-ID systems face the domain shift issue that training and testing domains are not cons...
main
Computer Vision
10.1609/aaai.v35i4.16422
35
4
3128-3135
official
null
null
10.1609/aaai.v35i4.16383
PGNet: Real-time Arbitrarily-Shaped Text Spotting with Point Gathering Network
https://ojs.aaai.org/index.php/AAAI/article/view/16383
https://ojs.aaai.org/index.php/AAAI/article/download/16383/16190
[ "Pengfei Wang", "Chengquan Zhang", "Fei Qi", "Shanshan Liu", "Xiaoqiang Zhang", "Pengyuan Lyu", "Junyu Han", "Jingtuo Liu", "Errui Ding", "Guangming Shi" ]
The reading of arbitrarily-shaped text has received increasing research attention. However, existing text spotters are mostly built on two-stage frameworks or character-based methods, which suffer from either Non-Maximum Suppression (NMS), Region-of-Interest (RoI) operations, or character-level annotations. In this pap...
main
Computer Vision
10.1609/aaai.v35i4.16383
35
4
2782-2790
official
2104.05458
title_snapshot
10.1609/aaai.v35i4.16384
Dynamic Position-aware Network for Fine-grained Image Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16384
https://ojs.aaai.org/index.php/AAAI/article/download/16384/16191
[ "Shijie Wang", "Haojie Li", "Zhihui Wang", "Wanli Ouyang" ]
Most weakly supervised fine-grained image recognition (WFGIR) approaches predominantly focus on learning the discriminative details which contain the visual variances and position clues. The position clues can be indirectly learnt by utilizing context information of discriminative visual content. However, this will cau...
main
Computer Vision
10.1609/aaai.v35i4.16384
35
4
2791-2799
official
null
null
10.1609/aaai.v35i4.16385
Co-mining: Self-Supervised Learning for Sparsely Annotated Object Detection
https://ojs.aaai.org/index.php/AAAI/article/view/16385
https://ojs.aaai.org/index.php/AAAI/article/download/16385/16192
[ "Tiancai Wang", "Tong Yang", "Jiale Cao", "Xiangyu Zhang" ]
Object detectors usually achieve promising results with the supervision of complete instance annotations. However, their performance is far from satisfactory with sparse instance annotations. Most existing methods for sparsely annotated object detection either re-weight the loss of hard negative samples or convert the ...
main
Computer Vision
10.1609/aaai.v35i4.16385
35
4
2800-2808
official
2012.01950
title_snapshot
10.1609/aaai.v35i4.16386
Very Important Person Localization in Unconstrained Conditions: A New Benchmark
https://ojs.aaai.org/index.php/AAAI/article/view/16386
https://ojs.aaai.org/index.php/AAAI/article/download/16386/16193
[ "Xiao Wang", "Zheng Wang", "Toshihiko Yamasaki", "Wenjun Zeng" ]
This paper presents a new high-quality dataset for Very Important Person Localization (VIPLoc), named Unconstrained-7k. Generally, current datasets: 1) are limited in scale; 2) built under simple and constrained conditions, where the number of disturbing non-VIPs is not large, the scene is relatively simple, and the fa...
main
Computer Vision
10.1609/aaai.v35i4.16386
35
4
2809-2816
official
null
null
10.1609/aaai.v35i4.16387
Teacher Guided Neural Architecture Search for Face Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16387
https://ojs.aaai.org/index.php/AAAI/article/download/16387/16194
[ "Xiaobo Wang" ]
Knowledge distillation is an effective tool to compress large pre-trained convolutional neural networks (CNNs) or their ensembles into models applicable to mobile and embedded devices. However, with expected flops or latency, existing methods are hand-crafted heuristics. They propose to pre-define the target student ne...
main
Computer Vision
10.1609/aaai.v35i4.16387
35
4
2817-2825
official
null
null
10.1609/aaai.v35i4.16388
Deep Multi-Task Learning for Diabetic Retinopathy Grading in Fundus Images
https://ojs.aaai.org/index.php/AAAI/article/view/16388
https://ojs.aaai.org/index.php/AAAI/article/download/16388/16195
[ "Xiaofei Wang", "Mai Xu", "Jicong Zhang", "Lai Jiang", "Liu Li" ]
Recent years have witnessed the growing interest in disease severity grading, especially for ocular diseases based on fundus images. The existing grading methods are usually trained with high resolution (HR) images. However, the grading performance decreases a lot given low resolution (LR) images, which are common in p...
main
Computer Vision
10.1609/aaai.v35i4.16388
35
4
2826-2834
official
null
null
10.1609/aaai.v35i4.16389
Confidence-aware Non-repetitive Multimodal Transformers for TextCaps
https://ojs.aaai.org/index.php/AAAI/article/view/16389
https://ojs.aaai.org/index.php/AAAI/article/download/16389/16196
[ "Zhaokai Wang", "Renda Bao", "Qi Wu", "Si Liu" ]
When describing an image, reading text in the visual scene is crucial to understand the key information. Recent work explores the TextCaps task, i.e. image captioning with reading Optical Character Recognition (OCR) tokens, which requires models to read text and cover them in generated captions. Existing approaches fai...
main
Computer Vision
10.1609/aaai.v35i4.16389
35
4
2835-2843
official
2012.03662
title_snapshot
10.1609/aaai.v35i4.16390
Geodesic-HOF: 3D Reconstruction Without Cutting Corners
https://ojs.aaai.org/index.php/AAAI/article/view/16390
https://ojs.aaai.org/index.php/AAAI/article/download/16390/16197
[ "Ziyun Wang", "Eric A. Mitchell", "Volkan Isler", "Daniel D. Lee" ]
Single-view 3D object reconstruction is a challenging fundamental problem in machine perception, largely due to the morphological diversity of objects in the natural world. In particular, high curvature regions are not always represented accurately by methods trained with common set-based loss functions such as Chamfer...
main
Computer Vision
10.1609/aaai.v35i4.16390
35
4
2844-2851
official
2006.07981
title_snapshot
10.1609/aaai.v35i4.16391
C2F-FWN: Coarse-to-Fine Flow Warping Network for Spatial-Temporal Consistent Motion Transfer
https://ojs.aaai.org/index.php/AAAI/article/view/16391
https://ojs.aaai.org/index.php/AAAI/article/download/16391/16198
[ "Dongxu Wei", "Xiaowei Xu", "Haibin Shen", "Kejie Huang" ]
Human video motion transfer (HVMT) aims to synthesize videos that one person imitates other persons' actions. Although existing GAN-based HVMT methods have achieved great success, they either fail to preserve appearance details due to the loss of spatial consistency between synthesized and exemplary images, or generate...
main
Computer Vision
10.1609/aaai.v35i4.16391
35
4
2852-2860
official
2012.08976
title_snapshot
10.1609/aaai.v35i4.16392
Semantic Consistency Networks for 3D Object Detection
https://ojs.aaai.org/index.php/AAAI/article/view/16392
https://ojs.aaai.org/index.php/AAAI/article/download/16392/16199
[ "Wenwen Wei", "Ping Wei", "Nanning Zheng" ]
Detecting 3D objects from point clouds is a significant yet challenging issue in many applications. While most existing approaches seek to leverage geometric information of point clouds, few studies accommodate the inherent semantic characteristics of each point and the consistency between the geometric and semantic cu...
main
Computer Vision
10.1609/aaai.v35i4.16392
35
4
2861-2869
official
null
null
10.1609/aaai.v35i4.16393
Holistic Multi-View Building Analysis in the Wild with Projection Pooling
https://ojs.aaai.org/index.php/AAAI/article/view/16393
https://ojs.aaai.org/index.php/AAAI/article/download/16393/16200
[ "Zbigniew Wojna", "Krzysztof Maziarz", "Łukasz Jocz", "Robert Pałuba", "Robert Kozikowski", "Iason Kokkinos" ]
We address six different classification tasks related to fine-grained building attributes: construction type, number of floors, pitch and geometry of the roof, facade material, and occupancy class. Tackling such a remote building analysis problem became possible only recently due to growing large-scale datasets of urba...
main
Computer Vision
10.1609/aaai.v35i4.16393
35
4
2870-2878
official
2008.10041
title_snapshot
10.1609/aaai.v35i4.16394
Stereopagnosia: Fooling Stereo Networks with Adversarial Perturbations
https://ojs.aaai.org/index.php/AAAI/article/view/16394
https://ojs.aaai.org/index.php/AAAI/article/download/16394/16201
[ "Alex Wong", "Mukund Mundhra", "Stefano Soatto" ]
We study the effect of adversarial perturbations of images on the estimates of disparity by deep learning models trained for stereo. We show that imperceptible additive perturbations can significantly alter the disparity map, and correspondingly the perceived geometry of the scene. These perturbations not only affect t...
main
Computer Vision
10.1609/aaai.v35i4.16394
35
4
2879-2888
official
2009.10142
title_snapshot
10.1609/aaai.v35i4.16395
Generalising without Forgetting for Lifelong Person Re-Identification
https://ojs.aaai.org/index.php/AAAI/article/view/16395
https://ojs.aaai.org/index.php/AAAI/article/download/16395/16202
[ "Guile Wu", "Shaogang Gong" ]
Existing person re-identification (Re-ID) methods mostly prepare all training data in advance, while real-world Re-ID data are inherently captured over time or from different locations, which requires a model to be incrementally generalised from sequential learning of piecemeal new data without forgetting what is alrea...
main
Computer Vision
10.1609/aaai.v35i4.16395
35
4
2889-2897
official
null
null
10.1609/aaai.v35i4.16396
Decentralised Learning from Independent Multi-Domain Labels for Person Re-Identification
https://ojs.aaai.org/index.php/AAAI/article/view/16396
https://ojs.aaai.org/index.php/AAAI/article/download/16396/16203
[ "Guile Wu", "Shaogang Gong" ]
Deep learning has been successful for many computer vision tasks due to the availability of shared and centralised large-scale training data. However, increasing awareness of privacy concerns poses new challenges to deep learning, especially for human subject related recognition such as person re-identification (Re-ID)...
main
Computer Vision
10.1609/aaai.v35i4.16396
35
4
2898-2906
official
2006.04150
title_snapshot
10.1609/aaai.v35i4.16397
Region-aware Global Context Modeling for Automatic Nerve Segmentation from Ultrasound Images
https://ojs.aaai.org/index.php/AAAI/article/view/16397
https://ojs.aaai.org/index.php/AAAI/article/download/16397/16204
[ "Huisi Wu", "Jiasheng Liu", "Wei Wang", "Zhenkun Wen", "Jing Qin" ]
We present a novel deep learning model equipped with a new region-aware global context modeling technique for automatic nerve segmentation from ultrasound images, which is a challenging task due to (1) the large variation and blurred boundaries of targets, (2) the large amount of speckle noise in ultrasound images, and...
main
Computer Vision
10.1609/aaai.v35i4.16397
35
4
2907-2915
official
null
null
10.1609/aaai.v35i4.16398
Precise Yet Efficient Semantic Calibration and Refinement in ConvNets for Real-time Polyp Segmentation from Colonoscopy Videos
https://ojs.aaai.org/index.php/AAAI/article/view/16398
https://ojs.aaai.org/index.php/AAAI/article/download/16398/16205
[ "Huisi Wu", "Jiafu Zhong", "Wei Wang", "Zhenkun Wen", "Jing Qin" ]
We propose a novel convolutional neural network (ConvNet) equipped with two new semantic calibration and refinement approaches for automatic polyp segmentation from colonoscopy videos. While ConvNets set state-of-the-are performance for this task, it is still difficult to achieve satisfactory results in a real-time man...
main
Computer Vision
10.1609/aaai.v35i4.16398
35
4
2916-2924
official
null
null
10.1609/aaai.v35i4.16399
Graph-to-Graph: Towards Accurate and Interpretable Online Handwritten Mathematical Expression Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16399
https://ojs.aaai.org/index.php/AAAI/article/download/16399/16206
[ "Jin-Wen Wu", "Fei Yin", "Yan-Ming Zhang", "Xu-Yao Zhang", "Cheng-Lin Liu" ]
Recent handwritten mathematical expression recognition (HMER) approaches treat the problem as an image-to-markup generation task where the handwritten formula is translated into a sequence (e.g. LaTeX). The encoder-decoder framework is widely used to solve this image-to-sequence problem. However, (i) for structured mat...
main
Computer Vision
10.1609/aaai.v35i4.16399
35
4
2925-2933
official
null
null
10.1609/aaai.v35i4.16400
Learning Comprehensive Motion Representation for Action Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16400
https://ojs.aaai.org/index.php/AAAI/article/download/16400/16207
[ "Mingyu Wu", "Boyuan Jiang", "Donghao Luo", "Junchi Yan", "Yabiao Wang", "Ying Tai", "Chengjie Wang", "Jilin Li", "Feiyue Huang", "Xiaokang Yang" ]
For action recognition learning, 2D CNN-based methods are efficient but may yield redundant features due to applying the same 2D convolution kernel to each frame. Recent efforts attempt to capture motion information by establishing inter-frame connections while still suffering the limited temporal receptive field or hi...
main
Computer Vision
10.1609/aaai.v35i4.16400
35
4
2934-2942
official
2103.12278
title_snapshot
10.1609/aaai.v35i4.16401
MVFNet: Multi-View Fusion Network for Efficient Video Recognition
https://ojs.aaai.org/index.php/AAAI/article/view/16401
https://ojs.aaai.org/index.php/AAAI/article/download/16401/16208
[ "Wenhao Wu", "Dongliang He", "Tianwei Lin", "Fu Li", "Chuang Gan", "Errui Ding" ]
Conventionally, spatiotemporal modeling network and its complexity are the two most concentrated research topics in video action recognition. Existing state-of-the-art methods have achieved excellent accuracy regardless of the complexity meanwhile efficient spatiotemporal modeling solutions are slightly inferior in per...
main
Computer Vision
10.1609/aaai.v35i4.16401
35
4
2943-2951
official
2012.06977
title_snapshot