Does the workshop have any proceedings?
No, we will only accept extended abstracts.
You are invited to submit extended abstracts to the third edition of the International Workshop in Ego4D which will be held alongside CVPR 2023 in Vancouver.
These abstracts represent existing or ongoing work and will not be published as part of any proceedings. We welcome all works that focus within the Egocentric Domain, it is not necessary to use the Ego4D dataset within your work. We expect a submission may contain one or more of the following topics (this is a non-exhaustive list):
The length of the extended abstracts is 2-4 pages, including figures, tables, and references. We invite submissions of ongoing or already published work, as well as reports on demonstrations and prototypes. The 3rd international Ego4D workshop gives opportunities for authors to present their work to the egocentric community to provoke discussion and feedback. Accepted work will be presented as either an oral presentation (either virtual or in-person) or as a poster presentation. The review will be single-blind, so there is no need to anonymize your work, but otherwise will follow the format of the CVPR submissions, information can be found here. Accepted abstracts will not be published as part of a proceedings, so can be uploaded to ArXiv etc. and the links will be provided on the workshop’s webpage. The submission will be managed with the Ego4D&EPIC@CVPR2023 CMT website. Use the Extended Abstract on CMT.
| Challenge Begin | March 2023 |
| Challenge Deadline | 19 May 2023 |
| Challenge Report Deadline | 26 May 2023 |
| Extended Abstract Deadline | 10 May 2023 |
| Notification to Authors | 24 May 2023 |
| Extended Abstracts ArXiv Deadline | 9 June 2023 |
| Workshop Date | 19 June 2023 |
All dates are local to Vancouver's time, PST.
Workshop Location: Room West 111-112
| Time | Event | Authors |
|---|---|---|
| Session 1 - Chairs: Giovanni Maria Farinella and Michael Wray | ||
| 08:30-08:45 | Welcome and Introductions | |
| 08:45-09:15 | Keynote 1 - Andrea Vedaldi (Univeristy of Oxford, UK) | |
| 09:15-10:15 | Ego4D Challenges - First Session |
9:15 - 9.25 -- Opening presentation 9:25 - 9:45 -- Episodic Memory 9:45 - 10:00 -- AV & Social 10:00 - 10:15 -- Forecasting |
| 10:15-10:45 | Coffee Break | |
| Session 2 - Chairs: David Crandall and Mike Shou | ||
| 10:45-11:15 | Keynote 2 - Hyun Soo Park (University of Minnesota, US) | |
| 11:15-12:00 | Ego4D Challenges - Second Session |
In-person presentation: GroundNLQ @ Ego4D Natural Language Queries Challenge 2023. Authors: Zhijian Hou (City University of Hong Kong), Lei Ji (Microsoft), Difei Gao (NUS), Wanjun Zhong (Sun Yat-Sen University) , Kun Yan (Beihang University), Chao Li (Microsoft), W. K. Chan (City University of Hong Kong), Chong-Wah Ngo (Singapore Management University), Mike Zheng Shou (National University of Singapore), Nan Duan (Microsoft Research) STHG: Spatial-Temporal Heterogeneous Graph Learning for Advanced Audio-Visual Diarization. Authors: Kyle Min (Intel Labs) QuAVF: Quality-aware Audio-Visual Fusion for Ego4D Talking to Me Challenge. Authors: Hsi-Che Lin (National Taiwan University), Chien-Yi Wang (NVIDIA) Min-Hung Chen (NVIDIA), Szu-Wei Fu (NVIDIA), Yu-Chiang Frank Wang (National Taiwan University) Recorded presentation: Single-Stage Visual Query Localization. Authors: Hanwen Jiang (University of Texas at Austin), Kristen Grauman (University of Texas at Austin & Meta AI) EgoLoc: Revisiting 3D Object Localization from Egocentric Videos with Visual Queries. Authors: Jinjie Mai (KAUST), Abdullah Hamdi (KAUST), Silvio Giancola (KAUST), Chen Zhao (KAUST), Bernard Ghanem (KAUST) Action Sensitivity Learning for the Ego4D Episodic Memory Challenge 2023. Authors: Jiayi Shao (Zhejiang University), Xiaohan Wang (Zhejiang University), Ruijie Quan (Zhejiang University), Yi Yang (Zhejiang University) Palm: Predicting Actions through Language Models @ Ego4D Long-Term Action Anticipation Challenge 2023. Authors: Daoji Huang (ETH Zurich), Otmar Hilliges (ETH Zurich), Luc Van Gool (ETH Zurich), Xi Wang (ETH Zurich) Prize ceremony: 11:50 - 12:00 |
| 12:00-12:30 | Invited CVPR Papers - First Session | |
| 12:00-12:06 | Paper 1: Egocentric Auditory Attention Localization in Conversations | Authors: Fiona Ryan (Georgia Institute of Technology), Hao Jiang (Meta Reality Labs Research), Abhinav Shukla (Meta Reality Labs Research), James M. Rehg (Georgia Institute of Technology), Vamsi Krishna Ithapu (Meta Reality Labs Research) |
| 12:06-12:12 | Paper 2: Where is my Wallet? Modeling Object Proposal Sets for Egocentric Visual Query Localization | Authors: Mengmeng Xu (KAUST), Yanghao Li (Meta AI), Cheng-Yang Fu (Meta AI), Bernard Ghanem (KAUST), Tao Xiang (Meta AI), Juan-Manuel Pérez-Rúa (Meta AI) |
| 12:12-12:15 | Question/Answering | |
| 12:15-12:21 | Paper 3: Ego-Body Pose Estimation via Ego-Head Pose Estimation | Authors: Jiaman Li (Stanford University), Karen Liu (Stanford University), Jiajun Wu (Stanford University) |
| 12:21-12:27 | Paper 4: Learning Video Representations from Large Language Models | Authors: Yue Zhao (The University of Texas at Austin), Ishan Misra (FAIR, Meta AI), Philipp Krähenbühl (The University of Texas at Austin), Rohit Girdhar (FAIR, Meta AI) |
| 12:27-12:30 | Question/Answering | |
| 12:30-13:30 | Lunch Break | |
| Session Chairs: Dima Damen and David Fouhey | ||
| 13:30-14:00 | Keynote 3 - Suraj Nair (Stanford University, US) | |
| 14:00-14:45 | EPIC Challenges - First Session | |
| 14:45-15:15 | Accepted Abstracts - First Session |
In-person presentation: FineBio: A Fine-Grained Video Dataset of Biological Experiments with Hierarchical Annotations. Authors: Takuma Yagi (National Institute of Advanced Industrial Science and Technology)*; Misaki Ohashi (The University of Tokyo); Yifei Huang (The University of Tokyo); Ryosuke Furuta (The University of Tokyo); Shungo Adachi (National Cancer Center Research Institute); Toutai Mitsuyama (National Institute of Advanced Industrial Science and Technology); Yoichi Sato (University of Tokyo) StillFast: An End-to-End Approach for Short-Term Object Interaction Anticipation. Authors: Francesco Ragusa (University of Catania)*; Giovanni Maria Farinella (University of Catania); Antonino Furnari (University of Catania). MECCANO: A Multimodal Egocentric Dataset for Humans Behavior Understanding in the Industrial-like Domain. Authors: Francesco Ragusa (University of Catania)*; Antonino Furnari (University of Catania); Giovanni Maria Farinella (University of Catania). Recorded Presentations: Prompting Large Language Models to Reformulate Queries for Moment Localization. Authors: Wenfeng Yan (Fudan University)*; Shaoxiang Chen (Fudan University); Zuxuan Wu (Fudan University); Yu-Gang Jiang (Fudan University). Recorded Presentation Link (YouTube). An Overview of Challenges in Egocentric Text-Video Retrieval. Authors: Burak Satar (Nanyang Technological University)*; Hongyuan Zhu (Institute for Infocomm, Research Agency for Science, Technology and Research (A*STAR) Singapore); Hanwang Zhang (Nanyang Technological University); Joo-Hwee Lim (Institute for Infocomm Research). Recorded Presentation Link (YouTube). |
| 15:15-15:45 | Coffee Break | |
| Session Chairs: Antonino Furnari and Michael Wray | ||
| 15:45-16:15 | Keynote 4 - David Fouhey (NYU/University of Michigan, US) | |
| 16:15-16:45 | EPIC Challenges - Second Session | |
| 16:45-17:15 | Aria Datasets and Challenges - Updates and Announcement | |
| 17:15-17:45 | Accepted Abstracts - Second Session |
In-person presentation: Enhancing Transformer Backbone for Egocentric Video Action Segmentation. Authors: Sakib Reza (Northeastern University)*; Balaji Sundareshan (Northeastern University); Mohsen Moghaddam (Northeastern University); Octavia Camps (Northeastern University). Abstract's ArXiv Link. Situated Cameras, Situated Knowledges: Towards an Egocentric Epistemology for Computer Vision. Authors: Samuel P Goree (Indiana University)*; David Crandall (Indiana University). Recorded Presentations: Monitoring Parkinson's Disease Progression Through Egocentric Vision: A Precision Health Approach. Authors: Nevasini NA Sasikumar (PESU)*; Krishna Sri Ipsit Mantri (Indian Institute of Technology Bombay). Human Action Recognition in Egocentric Perspective Using 2D Object and Hands Pose. Authors: Wiktor Mucha (Vienna University of Technology, Computer Vision Lab). Recorded Presentation Link (YouTube). Abstract's ArXiv Link. |
| 17:45-18:15 | Invited CVPR Papers - Second Session | |
| 17:45-17:51 | Paper 5: ARCTIC: A Dataset for Dexterous Bimanual Hand-Object Manipulation | Authors: Zicong Fan (ETH Zürich, Switzerland), Omid Taheri (Max Planck Institute for Intelligent Systems, Tübingen, Germany), Dimitrios Tzionas (University of Amsterdam, the Netherlands), Muhammed Kocabas (Max Planck Institute for Intelligent Systems, Tübingen, Germany), Manuel Kaufmann (ETH Zürich, Switzerland), Michael J. Black (Max Planck Institute for Intelligent Systems, Tübingen, Germany), Otmar Hilliges (ETH Zürich, Switzerland) |
| 17:51-17:57 | Paper 6: Egocentric Audio-Visual Object Localization | Authors: Chao Huang (University of Rochester), Yapeng Tian (University of Rochester), Anurag Kumar (Meta Reality Labs Research), Chenliang Xu (University of Rochester) |
| 17:57-18:00 | Question/Answering | |
| 18:00-18:06 | Paper 7: Scene-aware Egocentric 3D Human Pose Estimation | Authors: Jian Wang (Max Planck Institute for Informatics), Diogo Luvizon (Max Planck Institute for Informatics), Weipeng Xu (Meta Reality Labs), Lingjie Liu (Max Planck Institute for Informatics), Kripasindhu Sarkar (Google), Christian Theobalt (Max Planck Institute for Informatics) |
| 18:06-18:12 | Paper 8: AssemblyHands: Towards Egocentric Activity Understanding via 3D Hand Pose Estimation | Authors: Takehiko Ohkawa (Meta Reality Labs; The University of Tokyo), Kun He (Meta Reality Labs), Fadime Sener (Meta Reality Labs), Tomas Hodan (Meta Reality Labs), Luan Tran (Meta Reality Labs), Cem Keskin (Meta Reality Labs). |
| 18:12-18:15 | Question/Answering | |
| 18:15-18:45 | Future Plans and Closing Argument | |
No, we will only accept extended abstracts.
The Ego4D challenges will open in February 2023.
