Research map / Video / Video generation
Controllable Egocentric Video Generation via Occlusion-Aware Sparse 3D Hand Joints
Chenyangguang Zhang, Botao Ye, Boqi Chen, Alexandros Delitzas, Fangjinhua Wang, Marc Pollefeys, Xi Wang
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 occlusion-aware propagation (3D depth-weighted feature propagation) on top of OCA (ablation) | Ours · Ego4D | MPJPE: 2.75 -> 2.10 | Improved |
| add occlusion-aware propagation on top of OCA (ablation) | Ours · Ego4D | FID: 80.14 -> 76.08 | Improved |
| add occlusion-aware propagation on top of OCA (ablation) | Ours · EgoDex | FVD: 198.86 -> 190.01 | Improved |
| add 3D geometric embeddings on top of OCA+OP (ablation) | Ours · Ego4D | FID: 76.08 -> 67.70 | Improved |
| add 3D geometric embeddings on top of OCA+OP (ablation) | Ours · Ego4D | MPJPE: 2.10 -> 1.42 | Improved |
| add 3D geometric embeddings on top of OCA+OP (ablation) | Ours · EgoDex | FVD: 190.01 -> 174.73 | Improved |
| add 3D geometric embeddings on top of OCA+OP (ablation) | Ours · Ego4D | FVD: 300.99 -> 259.99 | Improved |
| use sparse 3D hand joints instead of projected 2D skeleton pose control (WAN-Fun) | Ours · Ego4D | FVD: 303.54 -> 259.99 | Improved |
| use 3D joints instead of 2D point tracks (MotionStream) | Ours · Ego4D | FVD: 308.63 -> 259.99 | Improved |
| use 3D joints instead of 2D point tracks (WAN-Move*) | Ours · EgoDex | MPJPE: 5.71 -> 1.80 | Improved |
| train WAN-Move on egocentric data | WAN-Move · Ego4D | MPJPE: 9.11 -> 9.77 | No clear effect |
| train WAN-Move on egocentric data | WAN-Move · Ego4D | FVD: 452.40 -> 332.44 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- WAN-Move* — Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance (NeurIPS 2025) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- MEgoHand: Multimodal Egocentric Hand-Object Interaction Motion Generation (NeurIPS 2025 · similarity 0.89)
- ManiVideo: Generating Hand-Object Manipulation Video with Dexterous and Generalizable Grasping (CVPR 2025 · similarity 0.89)
- StableHand: Quality-Aware Flow Matching for World-Space Dual-Hand Motion Estimation from Egocentric Video (NeurIPS 2026 · similarity 0.88)
- ComPose: When to Trust Hands for Object Pose Tracking (NeurIPS 2026 · similarity 0.88)
- The Surprising Effectiveness of Video Diffusion Models for Hand Motion Reconstruction (NeurIPS 2026 · similarity 0.88)
- HOT3D: Hand and Object Tracking in 3D from Egocentric Multi-View Videos (CVPR 2025 · similarity 0.88)
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.
