Hi team—besides the videos, does EgoLife provide (1) per-camera intrinsics (fx, fy, cx, cy + distortion model), (2) sensor extrinsics and/or per-frame 6-DoF poses, and (3) metric depth or any metric-scale supervision (e.g., stereo baselines, VIO)? Having these would enable accurate metric 3D representations of spatial objects. If available, could you point to the file paths/formats and a minimal loading example? Thanks!
Hi team—besides the videos, does EgoLife provide (1) per-camera intrinsics (fx, fy, cx, cy + distortion model), (2) sensor extrinsics and/or per-frame 6-DoF poses, and (3) metric depth or any metric-scale supervision (e.g., stereo baselines, VIO)? Having these would enable accurate metric 3D representations of spatial objects. If available, could you point to the file paths/formats and a minimal loading example? Thanks!