Research map / 3D vision / 3D generation
EgoWorld: Translating Exocentric View to Egocentric View using Rich Exocentric Observations
Junho Park, Andrew Sangwoo Ye, Taein Kwon
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 text conditioning (to no-condition model) (ablation) | EgoWorld · H2O unseen objects | FID: 56.120 -> 44.240 | Improved |
| add 2D hand pose map conditioning (to no-condition model) (ablation) | EgoWorld · H2O unseen objects | FID: 56.120 -> 55.016 | Improved |
| add pose to text-conditioned model (ablation) | EgoWorld · H2O unseen objects | FID: 44.240 -> 41.334 | Improved |
| use LDM inpainting backbone instead of MAT (ablation) | EgoWorld · H2O unseen objects | FID: 89.933 -> 41.334 | Improved |
| use LDM instead of MAE backbone (ablation) | EgoWorld · H2O unseen objects | FID: 169.91 -> 41.334 | Improved |
| add depth estimator and sparse map (ablation) | EgoWorld · H2O unseen actions | FID: 71.461 -> 42.323 | Improved |
| add 3D egocentric hand pose estimator (ablation) | EgoWorld · H2O unseen actions | FID: 62.714 -> 42.323 | Improved |
| use ViT backbone instead of CNN for egocentric hand pose estimator (ablation) | EgoWorld · H2O unseen actions | FID: 61.162 -> 42.323 | Improved |
| use hand pose rather than whole-body pose for transformation (ablation) | EgoWorld · H2O unseen actions | FID: 86.542 -> 42.323 | Improved |
| use MANO parameters instead of keypoints for hand pose (ablation) | EgoWorld · H2O unseen actions | FID: 33.284 -> 33.208 | No clear effect |
| use ground-truth hand pose and depth instead of predictions (ablation) | EgoWorld · H2O unseen actions | FID: 42.323 -> 33.284 | Improved |
| add video-to-video extension reusing previous-frame latent (ablation) | EgoWorld · H2O unseen actions | FID / temporal consistency: T-LPIPS 0.9827 -> 0.9860; FID 33.284 -> 35.455 | Mixed |
Closest papers by meaning
- EgoGen: An Egocentric Synthetic Data Generator (CVPR 2024 · similarity 0.91)
- Grounded-Exo2Ego: Structured Semantic Grounding for Robust Exocentric-to-Egocentric Video Generation (NeurIPS 2026 · similarity 0.89)
- Exocentric-to-Egocentric Video Generation (NeurIPS 2024 · similarity 0.88)
- Gaze Beyond the Frame: Forecasting Egocentric 3D Visual Span (NeurIPS 2025 · similarity 0.88)
- 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)
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.
