[ECCV 2026] CoMind: Understanding Collaborative Human Activity From Multiple Minds And Views

We introduce a novel egocentric and exocentric dataset capturing real-world collaboration in cooking scenarios. The dataset integrates time-synchronized multi-perspective video, high-quality audio, gaze tracking, and 3D scene and object scans, with annotations for shared object attention, social cues and agent-agent/agent-object interactions. More at: https://comind.ethz.ch/ Authors: Alexey Gavryushin *, Dingxi Zhang *, Zhao Huang *, Alexandros Delitzas, Jiaqi Chen, Ben Ellis, Cedric Zöllner, Manthan Patel, Manuel Kaufmann, Marc Pollefeys, Xi Wang.