Research map / Video / Video segmentation and tracking
Put Myself in Your Shoes: Lifting the Egocentric Perspective from Exocentric Videos
Mi Luo, Zihui Xue, Alex Dimakis, Kristen Grauman
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 |
|---|---|---|---|
| decouple structure (ego hand layout) prediction from diffusion pixel synthesis | Exo2Ego · H2O (new actions) | FID: 85.76 -> 38.03 | Improved |
| use Exo2Ego instead of best baseline on hand feasibility | Aria Pilot (new actions) · Aria Pilot (new actions) | Feasibility: 0.2646 -> 0.5869 | Improved |
| use Exo2Ego on unseen subject | Exo2Ego · H2O (unseen subject) | FID: 308.52 -> 150.87 | Improved |
| use geometry-aware pixelNeRF with camera parameters instead of generative approach | exo-to-ego translation · H2O (new actions) | FID: 38.03 -> 470.13 | Hurt |
| Exo2Ego versus vid2vid on SSIM for unseen object | Exo2Ego · H2O (unseen object) | SSIM: 0.349 -> 0.332 | No clear effect |
| choose exo cam 2 (widest FOV) rather than other cameras as source view (ablation) | H2O pipeline · H2O source exo camera | FID: 143.21 -> 132.03 | Improved |
| generalize to unseen scenes (background change) | Exo2Ego · H2O (unseen scene) | FID: 249.04 vs 325.34 (P-GAN) | Mixed |
Closest papers by meaning
- Exocentric-to-Egocentric Video Generation (NeurIPS 2024 · similarity 0.86)
- Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives (CVPR 2024 · similarity 0.85)
- Retrieval-Augmented Egocentric Video Captioning (CVPR 2024 · similarity 0.84)
- Object-Shot Enhanced Grounding Network for Egocentric Video (CVPR 2025 · similarity 0.84)
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Videos Generation (NeurIPS 2025 · similarity 0.83)
- Grounded-Exo2Ego: Structured Semantic Grounding for Robust Exocentric-to-Egocentric Video Generation (NeurIPS 2026 · similarity 0.82)
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.
