Research map / 3D vision / 3D generation
Ego-InBetween: Generating Object State Transitions in Ego-Centric Videos
Mengmeng Ge, Takashi Isobe, Xu Jia, Yanan Sun, Zetong Yang, Weinong Wang, Dong Zhou, Dong Li, Huchuan Lu, Emad Barsoum
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 TransitionVLM to DC-Interp baseline (ablation) | EgoIn · Epic100 | FVD: 296.67 -> 261.78 | Improved |
| add TransitionVLM to DC-Interp baseline (ablation) | EgoIn · Bridge | FVD: 287.47 -> 252.02 | Improved |
| add Transition Conditioning on top of TransitionVLM (ablation) | EgoIn · Epic100 | FVD: 261.78 -> 232.10 | Improved |
| add Transition Conditioning on top of TransitionVLM (ablation) | EgoIn · Bridge | FVD: 252.02 -> 220.92 | Improved |
| add Object-aware Auxiliary Supervision on top of TVLM and TC (ablation) | EgoIn · Epic100 | FVD: 232.10 -> 215.27 | Improved |
| add Object-aware Auxiliary Supervision on top of TVLM and TC (ablation) | EgoIn · Bridge | FVD: 220.92 -> 191.09 | Improved |
| use variable K in [1,4] instead of fixed K=1 (ablation) | EgoIn · Epic100 | FVD: 247.41 -> 215.27 | Improved |
| use variable K in [1,4] instead of fixed K=2 (ablation) | EgoIn · Epic100 | FVD: 228.66 -> 215.27 | Improved |
| use K=2 instead of K=1 (ablation) | EgoIn · Epic100 | FVD: 247.41 -> 228.66 | Improved |
| use K=4 instead of K=2 (ablation) | EgoIn · Epic100 | FVD: 228.66 -> 233.25 | Hurt |
| use K=4 instead of K=2 (ablation) | EgoIn · EgoFHO | FVD: 227.91 -> 230.90 | Hurt |
| use tuned TransitionVLM instead of original VLM for transition steps (ablation) | EgoIn · qualitative | visual realism | Improved |
Closest papers by meaning
- Generating 6DoF Object Manipulation Trajectories from Action Description in Egocentric Vision (CVPR 2025 · similarity 0.90)
- EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric View of Procedural Activities in Real World (CVPR 2024 · similarity 0.89)
- Action Scene Graphs for Long-Form Understanding of Egocentric Videos (CVPR 2024 · similarity 0.89)
- Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs (NeurIPS 2026 · similarity 0.88)
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Videos Generation (NeurIPS 2025 · similarity 0.88)
- EgoGen: An Egocentric Synthetic Data Generator (CVPR 2024 · 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.
