Research map / Video / Video generation
WorldWander: Bridging Egocentric and Exocentric Worlds in Video Generation
Quanjian Song, Yiren Song, Kelly Peng, Yuan Gao, Mike Zheng Shou
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 collaborative attention (token concatenation) instead of channel concatenation (ablation) | WorldWander · synthetic ego-to-exo | FVD: 524.3 -> 353.7 | Improved |
| add collaborative attention instead of channel concatenation (ablation) | WorldWander · synthetic exo-to-ego | FVD: 280.2 -> 230.3 | Improved |
| add collaborative attention instead of channel concatenation (ablation) | WorldWander · synthetic ego-to-exo | CLIP-I: 0.7310 -> 0.7672 | Improved |
| add different noise levels for condition and target instead of uniform timestep (ablation) | WorldWander · synthetic exo-to-ego | FVD: 287.7 -> 230.3 | Improved |
| add collaborative position encoding instead of uniform position encoding (ablation) | WorldWander · synthetic exo-to-ego | FVD: 280.3 -> 230.3 | Improved |
| add collaborative position encoding instead of uniform position encoding (ablation) | WorldWander · synthetic ego-to-exo | CLIP-I: 0.7562 -> 0.7672 | Improved |
| use LoRA rank 80 instead of 64 (ablation) | WorldWander · synthetic ego-to-exo | LPIPS: 0.5471 -> 0.5550 (LPIPS), 0.7461 -> 0.7672 (CLIP-I) | Mixed |
| use LoRA rank 80 instead of 128 (ablation) | WorldWander · synthetic exo-to-ego | FVD: 273.0 -> 230.3 | Improved |
| fine-tune ReCamMaster on EgoExo-8K | ReCamMaster · synthetic exo-to-ego | FVD: 738.7 -> 298.9 | Improved |
| use WorldWander instead of fine-tuned ReCamMaster | video translation · real-world ego-to-exo | FVD: 968.1 -> 374.7 | 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.
- ReCamMaster — ReCamMaster: Camera-Controlled Generative Rendering from A Single Video (ICCV 2025) · better on 2 of 2 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
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Videos Generation (NeurIPS 2025 · similarity 0.90)
- PlayerOne: Egocentric World Simulator (NeurIPS 2025 · similarity 0.89)
- Grounded-Exo2Ego: Structured Semantic Grounding for Robust Exocentric-to-Egocentric Video Generation (NeurIPS 2026 · similarity 0.88)
- Exocentric-to-Egocentric Video Generation (NeurIPS 2024 · similarity 0.88)
- EgoGen: An Egocentric Synthetic Data Generator (CVPR 2024 · similarity 0.87)
- WorldSimBench: Towards Video Generation Models as World Simulators (ICML 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.
