awesome-egocentric-vision

A curated list of egocentric (first-person) vision and related area resources

335

GitHub Stars

445

Curated Resources

Use this list with your AI agent

Add the Context Awesome MCP server to Claude, Cursor, or any MCP client, then ask:

"Show me eccv resources from awesome-egocentric-vision"

3D Hand Pose Estimation in Everyday Egocentric ImagesECCV
Aditya Prakash, Ruisen Tu, Matthew Chang, and Saurabh Gupta. In ECCV 2024.
3D Human Pose Perception from Egocentric Stereo VideosCVPR
Hiroyasu Akada, Jian Wang, Vladislav Golyanik, and Christian Theobalt. In CVPR 2024.
4Diff: 3D-Aware Diffusion Model for Third-to-First Viewpoint TranslationECCV
Feng Cheng, Mi Luo, Huiyu Wang, Alex Dimakis, Lorenzo Torresani, Gedas Bertasius, and Kristen Grauman. In ECCV 2024.
A Backpack Full of Skills: Egocentric Video Understanding with Diverse Task PerspectivesCVPR
Simone Alberto Peirone, Francesca Pistilli, Antonio Alliegro, and Giuseppe Averta. In CVPR 2024.
Action2Sound: Ambient-Aware Generation of Action Sounds from Egocentric VideosECCV
Changan Chen, Puyuan Peng, Ami Baid, Zihui Xue, Wei-Ning Hsu, David Harwath, and Kristen Grauman. In ECCV 2024.
Action Scene Graphs for Long-Form Understanding of Egocentric VideosCVPR
Ivan Rodin, Antonino Furnari, Kyle Min, Subarna Tripathi, and Giovanni Maria Farinella. In CVPR 2024.

ADLAll Datasets
20 subjects performing daily activities in their native environments.
Aria Digital TwinAll Datasets
A comprehensive egocentric dataset containing 200 sequences of real-world activities conducted by Aria wearers in two real indoor scenes with 398 object instances (324 stationary and 74 dynamic).
Aria Everyday Activities (AEA)All Datasets
143 daily-activity sequences recorded by multiple wearers in five indoor locations with Project Aria, with globally-aligned 3D trajectories, scene point clouds, per-frame eye gaze, and time-aligned speech.
Aria Gen 2 Pilot Dataset (A2PD)All Datasets
Egocentric multimodal dataset captured with Aria Gen 2 glasses across everyday scenarios (cleaning, cooking, eating, playing, walking) between a primary user and three friends, with rich sensor streams (RGB, eye tracking, IMU, spatial audio, PPG, GPS) and machine-perception outputs.
Aria Navigation Dataset (AND)All Datasets
Roughly 4 hours of Project Aria egocentric recordings for humanoid/egocentric navigation, introduced with LookOut.
Assembly101All Datasets
4,321 videos of assembling/disassembling 101 take-apart toy vehicles, with 8 static and 4 egocentric views, 100K+ coarse and 1M fine-grained action segments, and 18M 3D hand poses, for procedural activity recognition, anticipation, segmentation, and mistake detection.

Showing a sample of 445 resources. View the full list on GitHub →