Research map / Video / Video generation
From Synchrony to Sequence: Exo-to-Ego Generation via Interpolation
Mohammad Mahdi, Nedko Savov, Danda Paudel, Luc Van Gool
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 frame interpolation (no pose interpolation) (ablation) | Syn2Seq-Forcing · Ego-Exo4D Health | PSNR: 13.5419 -> 16.1003 | Improved |
| add frame interpolation (no pose interpolation) (ablation) | Syn2Seq-Forcing · Ego-Exo4D Bike | PSNR: 14.0091 -> 15.1245 | Improved |
| add frame interpolation (no pose interpolation) (ablation) | Syn2Seq-Forcing · Ego-Exo4D Cooking | PSNR: 13.1609 -> 14.2131 | Improved |
| add frame interpolation (no pose interpolation) (ablation) | Syn2Seq-Forcing · Ego-Exo4D Health | LPIPS: 0.5822 -> 0.4902 | Improved |
| add pose interpolation on top of frame interpolation (ablation) | Syn2Seq-Forcing · Ego-Exo4D Health | PSNR: 16.1003 -> 16.7139 | Improved |
| add pose interpolation on top of frame interpolation (ablation) | Syn2Seq-Forcing · Ego-Exo4D Bike | SSIM: 0.4405 -> 0.4721 | Improved |
| add pose interpolation on top of frame interpolation (ablation) | Syn2Seq-Forcing · Ego-Exo4D Cooking | PSNR: 14.2131 -> 14.3897 | Improved |
| use WFLF interpolator instead of DFoT native interpolation (ablation) | Syn2Seq-Forcing · Ego-Exo4D Bike (both segments) | PSNR: 13.4801 -> 15.6522 | Improved |
| use WFLF interpolator instead of DFoT native interpolation (ablation) | Syn2Seq-Forcing · Ego-Exo4D Bike (both segments) | LPIPS: 0.6258 -> 0.5016 | Improved |
| use Plucker pose embedding instead of Global (ablation) | Syn2Seq-Forcing · Ego-Exo4D Bike (ego segment) | PSNR: 15.0197 -> 15.6301 | Improved |
| use Plucker pose embedding instead of Ray Encoding (ablation) | Syn2Seq-Forcing · Ego-Exo4D Bike (ego segment) | PSNR: 15.5702 -> 15.6301 | Improved |
| use Plucker pose embedding instead of Ray Encoding (ablation) | Syn2Seq-Forcing · Ego-Exo4D Bike (ego segment) | SSIM: 0.4729 -> 0.4721 | Mixed |
Closest papers by meaning
- Exocentric-to-Egocentric Video Generation (NeurIPS 2024 · similarity 0.90)
- Grounded-Exo2Ego: Structured Semantic Grounding for Robust Exocentric-to-Egocentric Video Generation (NeurIPS 2026 · similarity 0.89)
- Bootstrap Your Own Views: Masked Ego-Exo Modeling for Fine-grained View-invariant Video Representations (CVPR 2025 · similarity 0.87)
- AR-Diffusion: Asynchronous Video Generation with Auto-Regressive Diffusion (CVPR 2025 · similarity 0.87)
- SynCamMaster: Synchronizing Multi-Camera Video Generation from Diverse Viewpoints (ICLR 2025 · similarity 0.87)
- EgoExo-Gen: Ego-centric Video Prediction by Watching Exo-centric Videos (ICLR 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.
