Research map / 3D vision / 3D generation
EgoSim: Egocentric World Simulator for Embodiment Interaction Generation
Jinkun Hao, Mingda Jia, Xudong Xu, Ruiyan Wang, Xihui Liu, Ran Yi, Lizhuang Ma, Jiangmiao Pang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add camera-trajectory rendering of the point cloud (ablation) | EgoSim · EgoDex (tabletop) | PSNR: 23.380 -> 25.056 | Improved |
| add camera-trajectory rendering of the point cloud (ablation) | EgoSim · EgoDex (tabletop) | Depth-ERR: 10.238 -> 8.888 | Improved |
| add accurate visibility mask instead of all-black mask (ablation) | EgoSim · EgoDex (tabletop) | PSNR: 23.988 -> 25.056 | Improved |
| add accurate visibility mask instead of all-black mask (ablation) | EgoSim · EgoDex (tabletop) | Depth-ERR: 14.124 -> 8.888 | Improved |
| pretrain on egocentric human hand data before robot finetuning (ablation) | EgoSim · AgiBot robot | PSNR: 16.36 -> 18.67 | Improved |
| pretrain on egocentric human hand data before robot finetuning (ablation) | EgoSim · AgiBot robot | LPIPS: 0.31 -> 0.28 | Improved |
| pretrain on egocentric human hand data before robot finetuning (ablation) | EgoSim · AgiBot robot | SSIM: 0.69 -> 0.72 | Improved |
| continuous generation with intermediate 3D state update vs single clip | EgoSim · EgoDex (tabletop) | Depth-ERR: 8.888 -> 10.943 | Hurt |
| continuous generation with intermediate 3D state update vs single clip | EgoSim · EgoDex (tabletop) | Cam-ERR: 0.0013 -> 0.0017 | Hurt |
| replace InterDyn 2D hand mask conditioning with explicit 3D point-cloud conditioning | InterDyn · TRUMANS | Depth-err (GT): 57.962 -> 16.084 | Improved |
| replace InterDyn with EgoSim on hand keypoints | InterDyn · EgoDex / EgoVid | PCK@20: 0.6481 -> 0.5141 (EgoDex); 0.5962 -> 0.6122 (EgoVid) | Mixed |
| replace InterDyn with EgoSim on object 2D AP | InterDyn · EgoDex | 2D AP: 6.53 -> 21.98 | Improved |
Closest papers by meaning
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Videos Generation (NeurIPS 2025 · similarity 0.90)
- PlayerOne: Egocentric World Simulator (NeurIPS 2025 · similarity 0.90)
- EgoGen: An Egocentric Synthetic Data Generator (CVPR 2024 · similarity 0.90)
- EgoSim: An Egocentric Multi-view Simulator and Real Dataset for Body-worn Cameras during Motion and Activity (NeurIPS 2024 · similarity 0.88)
- EgoChoir: Capturing 3D Human-Object Interaction Regions from Egocentric Views (NeurIPS 2024 · similarity 0.88)
- MEgoHand: Multimodal Egocentric Hand-Object Interaction Motion Generation (NeurIPS 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
