Dataset Viewer
Auto-converted to Parquet Duplicate
image
imagewidth (px)
1.15k
1.7k
text
stringlengths
190
6.34k
# V2X-Sim: Multi-Agent Collaborative Perception Dataset and Benchmark for Autonomous Driving Yiming \(\mathrm{Li}^{1}\) , Dekun \(\mathrm{Ma}^{1}\) , Ziyan \(\mathrm{An}^{1}\) , Zixun Wang \(^{1}\) , Yiqi Zhong \(^{2}\) , Siheng Chen \(^{3,\infty}\) , and Chen Feng \(^{1,\infty}\) Abstract—Vehicle- to- everything...
# The Devil Is in the Details: Window-based Attention for Image Compression Renjie Zou 1 Chunfeng Song 1 Zhaoxiang Zhang 1,2,\\* 1 National Laboratory of Pattern Recognition (NLPR) of Institute of Automation, Chinese Academy of Sciences (CASIA) & University of Chinese Academy of Sciences (UCAS) 2 Centre for...
# Shape-guided Object Inpainting Yu Zeng \(^{1}\) , Zhe Lin \(^{2}\) , and Vishal M. Patel \(^{1}\) \(^{1}\) Johns Hopkins University, {yzeng22, vpatel36}@jhu.edu \(^{2}\) Adobe Research zlin@adobe.com <|ref|>image<|/ref|><|det|>[[77, 243, 918, 454]]<|/det|> <center>Fig. 1: We explore a new image inpainting ta...
# Cylin-Painting: Seamless \(360^{\circ}\) Panoramic Image Outpainting and Beyond Kang Liao\*, Xiangyu \(\mathrm{Xu^{*,\dagger}}\) , Chunyu Lin†, Wenqi Ren, Yunchao Wei, Yao Zhao, Fellow, IEEE Abstract- Image outpainting gains increasing attention since it can generate the complete scene from a partial view, prov...
# PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models Yuan Yao \(^{1*}\) , Qianyu Chen \(^{1*}\) , Ao Zhang \(^{2}\) , Wei Ji \(^{2}\) Zhiyuan Liu \(^{1\dagger}\) , Tat- Seng Chua \(^{2}\) , Maosong Sun \(^{1\dagger}\) \(^{1}\) Dept. of Comp. Sci. & Tech., Institute for AI, Tsinghua Uni...
# Contextual Text Block Detection towards Scene Text Understanding Chuhui Xue \(^{1,2}\) , Jiaxing Huang \(^{1}\) , Shijian Lu \(^{1}\) , Changhu Wang \(^{2}\) , and Song Bai \(^{2}\) \(^{1}\) Nanyang Technological University \(^{2}\) ByteDance Inc. xuec0003@e.ntu.edu.sg, {jiaxing.huang,shijian.lu}@ntu.edu.sg, {c...
# Exploring Adversarial Robustness of Vision Transformers in the Spectral Perspective Gihyun Kim Juyeop Kim Jong- Seok Lee Yonsei University, Republic of Korea {kkh9314, juyeopkim, jong- seok.lee}@yonsei.ac.kr Abstract—The Vision Transformer has emerged as a powerful tool for image classification tasks, surpassin...
# Long-Tailed Class Incremental Learning Xialei Liu \(^{1, *}\) , Yu- Song Hu \(^{1, *}\) , Xu- Sheng Cao \(^{1}\) , Andrew D. Bagdanov \(^{2}\) , Ke Li \(^{3}\) , and Ming- Ming Cheng \(^{1}\) TMCC, CS, Nankai University, China \(^{1}\) ; MICC, University of Florence, Florence, Italy \(^{2}\) Tencent Youtu Lab \...
# DE-FAKE: Detection and Attribution of Fake Images Generated by Text-to-Image Generation Models Zeyang Sha<sup>1</sup> Zheng Li<sup>1</sup> Ning Yu<sup>2</sup> Yang Zhang<sup>1</sup> <sup>1</sup>CISPA Helmholtz Center for Information Security <sup>2</sup> Salesforce Research ## Abstract Text- to- image gen...
# MagicVideo: Efficient Video Generation With Latent Diffusion Models Daquan Zhou\* Weimin Wang\* Hanshu Yan Weiwei Lv Yizhe Zhu Jiashi Feng ByteDance Inc. {daquanzhou, weimin.wang, hanshu.yan}@bytedance.com {vici, yizhe.zhu, jshfeng}@bytedance.com ## Abstract We present an efficient text- to- video gene...
# LoCoNet: Long-Short Context Network for Active Speaker Detection Xizi Wang \(^{1}\) Feng Cheng \(^{2}\) Gedas Bertasius \(^{2}\) David Crandall \(^{1}\) \(^{1}\) Indiana University \(^{2}\) UNC Chapel Hill xiziwang@iu.edu {fengchan,gedas}@cs.unc.edu ## Abstract Active Speaker Detection (ASD) aims to identify...
# Composer: Creative and Controllable Image Synthesis with Composable Conditions Lianghua Huang' Di Chen' Yu Liu' Yujun Shen' Deli Zhao' Jingren Zhou' ## Abstract Recent large- scale generative models learned on big data are capable of synthesizing incredible images yet suffer from limited controllability. Thi...
# Traffic Scene Parsing through the TSP6K Dataset Peng- Tao Jiang \(^{1,2*}\) Yuqi Yang \(^{1*}\) Yang Cao \(^{3}\) Qibin Hou \(^{1,4\dagger}\) Ming- Ming Cheng \(^{1,4}\) Chunhua Shen \(^{2}\) \(^{1}\) VCIP, CS, Nankai University \(^{2}\) Zhejiang University \(^{3}\) HKUST \(^{4}\) NKIARI, Shenzhen Futian pt.jiang@...
# 3DInAction: Understanding Human Actions in 3D Point Clouds Yizhak Ben- Shabat \(^{1,2}\) Oren Shrout \(^{2}\) Stephen Gould \(^{1}\) \(^{1}\) Australian National University \(^{2}\) Technion, Israel Institute of Technology sitzikbs@technion.ac.il, shrout.oren@campus.technion.ac.il, stephen.gould@anu.edu.au http...
# Unifying Top-down and Bottom-up Scanpath Prediction Using Transformers Zhibo Yang \(^{1,2}\) , Sounak Mondal \(^{1}\) , Seoyoung Ahn \(^{1}\) , Ruoyu Xue \(^{1}\) , Gregory Zelinsky \(^{1}\) , Minh Hoai \(^{1,3}\) , Dimitris Samaras \(^{1}\) \(^{1}\) Stony Brook University \(^{2}\) Waymo LLC \(^{3}\) VinAI Researc...
# MoRF: Mobile Realistic Fullbody Avatars from a Monocular Video Renat Bashirov \(^{1}\) Alexey Larionov \(^{1}\) Evgeniya Ustinova \(^{1}\) Mikhail Sidorenko \(^{1}\) David Svitov \(^{1}\) Ilya Zakharkin \(^{2*}\) Victor Lempitsky \(^{3*}\) \(^{1}\) Samsung Research \(^{2}\) ZERO10 \(^{3}\) Cinemersive Labs ## A...
# SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude Operations Pu Li \(^{1,2}\) Jianwei Guo \(^{1,2*}\) Xiaopeng Zhang \(^{1,2}\) Dong- Ming Yan \(^{1,2}\) \(^{1}\) MAIS, Institute of Automation, Chinese Academy of Sciences \(^{2}\) School of Artificial Intelligence, University of Chinese Acad...
# Non-Exemplar Online Class-incremental Continual Learning via Dual-prototype Self-augment and Refinement —Appendix— Fushuo Huo \(^{1}\) , Wenchao Xu \(^{1}\) , Jingcai Guo \(^{1,2}\) , Haozhao Wang \(^{3,*}\) , Yunfeng Fan \(^{1}\) \(^{1}\) Department of Computing, The Hong Kong Polytechnic University, China \(^...
# SpikeCV: Open a Continuous Computer Vision Era Yajing Zheng, Jiyuan Zhang, Rui Zhao, Jianhao Ding, Shiyan Chen, Ruiqin Xiong, Zhaofei Yu, and Tiejun Huang Abstract—SpikeCV is a new open- source computer vision platform for the spike camera, which is a neuromorphic visual sensor that has developed rapidly in rec...
# Exploring Efficient Asymmetric Blind-Spots for Self-Supervised Denoising in Real-World Scenarios Shiyan Chen \(^{1,2}\) Jiyuan Zhang \(^{1,2}\) Zhaofei Yu \(^{1,2,3*}\) Tiejun Huang \(^{1,2,3}\) \(^{1}\) School of Computer Science, Peking University \(^{2}\) National Key Laboratory for Multimedia Information Proce...
# Overload: Latency Attacks on Object Detection for Edge Devices Erh- Chung Chen \(^{1}\) \* Pin- Yu Chen \(^{2}\) I- Hsin Chung \(^{2}\) Che- Rung Lee \(^{1}\) National Tsing Hua University \(^{1}\) IBM Research \(^{2}\) ## Abstract Nowadays, the deployment of deep learning- based applications is an essential...
# DETRs Beat YOLOs on Real-time Object Detection Yian Zhao \(^{1,2\dagger}\) Wenyu Lv \(^{1\dagger \ddagger}\) Shangliang Xu \(^{1}\) Jinman Wei \(^{1}\) Guanzhong Wang \(^{1}\) Qingqing Dang \(^{1}\) Yi Liu \(^{1}\) Jie Chen \(^{2\boxtimes}\) \(^{1}\) Baidu Inc, Beijing, China \(^{2}\) School of Electronic and C...
# TransFlow: Transformer as Flow Learner Yawen Lu \(^{1,2}\) , Qifan Wang \(^{3}\) , Siqi Ma \(^{4}\) , Tong Geng \(^{5}\) , Yingjie Victor Chen \(^{1}\) , Huaijin Chen \(^{6}\) , and Dongfang Liu \(^{2*}\) \(^{1}\) Purdue University \(^{2}\) Rochester Institute of Technology \(^{3}\) Meta AI \(^{4}\) Westlake Un...
# Enhancing Video Super-Resolution via Implicit Resampling-based Alignment Kai Xu Ziwei Yu Xin Wang Michael Bi Mi Angela Yao kxu@comp.nus.edu.sg ## Abstract In video super- resolution, it is common to use a frame- wise alignment to support the propagation of information over time. The role of alignment is well...
# ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding Le Xue \(^{1*}\) Ning Yu \(^{1}\) Shu Zhang \(^{1}\) Artemis Panagopoulou \(^{1,3}\) Junnan Li \(^{1}\) Roberto Martin- Martin \(^{4}\) Jiajun Wu \(^{2}\) Caiming Xiong \(^{1}\) Ran Xu \(^{1}\) Juan Carlos Niebles \(^{1,2}\) Silvio Savarese \(^{...
# One-Prompt to Segment All Medical Images Junde Wu National University of Singapore MBZUAI University of Oxford Jiayuan Zhu University of Oxford Yuanpei Liu University of Hongkong Yueming Jin National University of Singapore Min Xu Carnegie Mellon University MBZUAI ## Abstract Large foundation mod...
# Overcoming Topology Agnosticism: Enhancing Skeleton-Based Action Recognition through Redefined Skeletal Topology Awareness Yuxuan Zhou \(^{1}\) Zhi- Qi Cheng \(^{2*}\) Jun- Yan He \(^{3}\) Bin Luo \(^{3}\) Yifeng Geng \(^{3}\) Xuansong Xie \(^{3}\) \(^{1}\) University of Mannheim, \(^{2}\) Carnegie Mellon Universi...
# RoMa: Robust Dense Feature Matching Johan Edstedt<sup>1</sup> Qiyu Sun<sup>2</sup> Georg Bokman<sup>3</sup> Mårten Wadenbäck<sup>1</sup> Michael Felsberg<sup>1</sup> <sup>1</sup>Linköping University, <sup>2</sup>East China University of Science and Technology, <sup>3</sup>Chalmers University of Technology <|re...
# Zero-TPrune: Zero-Shot Token Pruning through Leveraging of the Attention Graph in Pre-Trained Transformers Hongjie Wang, Bhishma Dedhia, Niraj K. Jha Princeton University Princeton, NJ 08540, USA {hongjiewang, bdedhia, jha}@princeton.edu ## Abstract Deployment of Transformer models on edge devices is beco...
# StyleGAN knows Normal, Depth, Albedo, and More Anand Bhattad Daniel McKee Derek Hoiem D.A. Forsyth University of Illinois Urbana Champaign <|ref|>image<|/ref|><|det|>[[177, 280, 822, 459]]<|/det|> <center>Figure 1: StyleGAN has easily accessible and accurate representations of intrinsic images, without ever hav...
# ProTeCt: Prompt Tuning for Taxonomic Open Set Classification Tz- Ying Wu\* Chih- Hui Ho\* Nuno Vasconcelos University of California, San Diego {tzw001, chh279, nvasconcelos}@ucsd.edu ## Abstract Visual- language foundation models, like CLIP, learn generalized representations that enable zero- shot open- se...
# Training Like a Medical Resident: Context-Prior Learning Toward Universal Medical Image Segmentation Yunhe Gao \(^{1}\) Zhuowei Li \(^{1}\) Di Liu \(^{1}\) Mu Zhou \(^{1}\) Shaoting Zhang \(^{2}\) Dimitris Metaxas \(^{1}\) Rutgers University, \(^{2}\) Shanghai AI Laboratory ## Abstract A major focus of clini...
# MMSum: A Dataset for Multimodal Summarization and Thumbnail Generation of Videos Jielin Qiu \(^{1,2}\) , Jiacheng Zhu \(^{3}\) , William Han \(^{1}\) , Aditesh Kumar \(^{1}\) , Karthik Mittal \(^{1}\) , Claire Jin \(^{1}\) , Zhengyuan Yang \(^{2}\) , Linjie Li \(^{2}\) , Jianfeng Wang \(^{2}\) , Ding Zhao \(^{1}\)...
# CorrMatch: Label Propagation via Correlation Matching for Semi-Supervised Semantic Segmentation Boyuan Sun<sup>1</sup> Yu- Qi Yang<sup>1</sup> Le Zhang<sup>2</sup> Ming- Ming Cheng<sup>1</sup> Qibin Hou<sup>1\*</sup> <sup>1</sup>VCIP, School of Computer Science, Nankai University <sup>2</sup>School of Informat...
# Referring Expression Comprehension Using Language Adaptive Inference Wei Su \(^{1}\) , Peihan Miao \(^{2}\) , Huanzhang Dou \(^{1}\) , Yongjian Fu \(^{1}\) , Xi Li \(^{1,3,4*}\) \(^{1}\) College of Computer Science & Technology, Zhejiang University \(^{2}\) School of Software Technology, Zhejiang University \(^...
# WOUAF: Weight Modulation for User Attribution and Fingerprinting in Text-to-Image Diffusion Models Changhoon Kim\*1 Kyle Min\*2 Maitreya Patel1 Sheng Cheng1 Yezhou Yang1 1Arizona State University 2Intel Labs {kch,maitreya.patel,scheng53,yz.yang}@asu.edu kyle.min@intel.com ## Abstract The rapid advancement...
# FLEdge: Benchmarking Federated Machine Learning Applications in Edge Computing Systems Herbert Woisetschlager\* Technical University of Munich Germany herbert.woisetschlaeger@tum.de Alexander Isenko Technical University of Munich Germany alex.isenko@tum.de Ruben Mayer University of Bayreuth Germany ruben.may...
# Grounded Text-to-Image Synthesis with Attention Refocusing Quynh Phung Songwei Ge Jia- Bin Huang University of Maryland College Park https://attention- refocusing.github.io/ <|ref|>image<|/ref|><|det|>[[85, 245, 477, 345]]<|/det|> <center>Three apples are sitting side-by-side on a wooden table </center> <|re...
# Multi-Modal Classifiers for Open-Vocabulary Object Detection Pranmay Kaul \(^{1}\) Weidi Xie \(^{1,2,3}\) Andrew Zisserman \(^{1}\) https://www.robots.ox.ac.uk/vgg/research/mm- ovod/ ## Abstract The goal of this paper is open- vocabulary object detection (OVOD) — building a model that can detect objects beyo...
# ModeT: Learning Deformable Image Registration via Motion Decomposition Transformer Haiqiao Wang, Dong Ni, and Yi Wang Smart Medical Imaging, Learning and Engineering (SMILE) Lab, Medical UltraSound Image Computing (MUSIC) Lab, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen Univer...
# Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions Weizhen He \(^{1}\) Yiheng Deng \(^{1}\) Shixiang Tang \(^{2}\) Qihao Chen \(^{4}\) Qingsong Xie \(^{5}\) Yizhou Wang \(^{6}\) Lei Bai \(^{6}\) Feng Zhu \(^{3}\) Rui Zhao \(^{3,7}\) Wanli Ouyang \(^{2,6}\) Donglian Qi \(^{1}\) Yunfeng Y...
# Robustness of SAM: Segment Anything Under Corruptions and Beyond Yu Qiao, Chaoning Zhang, Taegoo Kang, Donghun Kim, Chenshuang Zhang, and Choong Seon Hong, Senior, IEEE Abstract—Segment anything model (SAM), as the name suggests, is claimed to be capable of cutting out any object and demonstrates impressive zer...
# Visual Language Pretrained Multiple Instance Zero-Shot Transfer for Histopathology Images Ming Y. Lu<sup>1,1,2,3</sup>, Bowen Chen<sup>1,2,3</sup>, Andrew Zhang<sup>1,2,3</sup>, Drew F.K. Williamson<sup>2,3</sup>, Richard J. Chen<sup>2,3</sup>, Tong Ding<sup>2,3</sup>, Long Phi Le<sup>2,3</sup>, Yung- Sung Chuang<...
# Efficient 3D Semantic Segmentation with Superpoint Transformer Damien Robert \(^{1,2}\) Hugo Raguet \(^{3}\) Loic Landrieu \(^{2,4}\) damien.robert@ign.fr hugo.raguet@insa- cvl.fr loic.landrieu@enpc.fr \(^{1}\) CSAI, ENGIE Lab CRIGEN, France \(^{2}\) LASTIG, IGN, ENSG, Univ Gustave Eiffel, France \(^{3}\) INSA ...
# LoSh: Long-Short Text Joint Prediction Network for Referring Video Object Segmentation Linfeng Yuan, Miaojing Shi\*, Zijie Yue, Qijun Chen College of Electronic and Information Engineering, Tongji University linfengyuan1997@gmail.com, {mshi,zijie,qjchen}@tongji.edu.cn ## Abstract Referring video object seg...
# Contrasting intra-modal and ranking cross-modal hard negatives to enhance visio-linguistic compositional understanding Le Zhang \(^{1,2}\) Rabiul Awal \(^{1}\) Aishwarya Agrawal \(^{1,2,3}\) Mila - Quebec AI Institute \(^{1}\) Université de Montréal \(^{2}\) Canada CIFAR AI Chair \(^{3}\) ## Abstract Vision-...
# Infinite Photorealistic Worlds using Procedural Generation Alexander Raistrick\*, Lahav Lipson\*, Zeyu Ma\*, (\*equal contribution; alphabetical order) Lingjie Mei, Mingzhe Wang, Yiming Zuo, Karhan Kayan, Hongyu Wen, Beining Han, Yihan Wang, Alejandro Newell†, Hei Law†, Ankit Goyal†, Kaiyu Yang†, Jia Deng Departme...
# ArtFusion: Controllable Arbitrary Style Transfer using Dual Conditional Latent Diffusion Models Dar- Yen Chen chendaryen@outlook.com <|ref|>image<|/ref|><|det|>[[86, 274, 888, 430]]<|/det|> <center>Figure 1: Results of our ArtFusion using classifier-free guidance along the style and content conditions. We can a...
# Generative Proxemics: A Prior for 3D Social Interaction from Images Lea Müller<sup>1</sup> Vickie Ye<sup>2</sup> Georgios Pavlakos<sup>2</sup> Michael Black<sup>1</sup> Angjoo Kanazawa<sup>2</sup> <sup>1</sup>MPI for Intelligent Systems, Tübingen, Germany <sup>2</sup>UC Berkeley {lea.mueller,black}@tuebingen.mp...
End of preview. Expand in Data Studio
def remove_ref_tags_except_image_from_sample(sample):
    """
    cleanup the deepseek exrtra tags
    """
    lines = sample["text"].split('\n')
    filtered_lines = []
    
    for line in lines:
        # Check if line contains any <|ref|> tag
        if '<|ref|>' in line:
            # Keep the line only if it contains <|ref|>image<|/ref|>
            if '<|ref|>image<|/ref|>' in line:
                filtered_lines.append(line)
        else:
            # Keep lines without <|ref|> tags
            filtered_lines.append(line)
    
    out = '\n'.join(filtered_lines)
    return {'text':out.strip()}
Downloads last month
19