Research map / Video / Video segmentation and tracking
Few-shot Video-to-Video Synthesis
Ting-Chun Wang, Ming-Yu Liu, Andrew Tao, Guilin Liu, Bryan Catanzaro, Jan Kautz
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 |
|---|---|---|---|
| Use example-conditioned weight generation instead of the baseline approach | Few-shot Video-to-Video Synthesis · video-to-video synthesis | rendering quality | Improved |
| Use the proposed model with one example image (K=1) | Few-shot Video-to-Video Synthesis · specific person (Table B) | rendering quality | No clear effect |
| Fine-tune the model on the example images | Few-shot Video-to-Video Synthesis · specific person (Table B) | rendering quality | Improved |
| Increase the number of example images | Few-shot Video-to-Video Synthesis · video-to-video synthesis | rendering quality | Improved |
| Use AdaIN rather than the proposed weight-generation scheme as dataset size grows | Few-shot Video-to-Video Synthesis · larger training datasets | performance | Hurt |
| Test on a domain very different from the training domains | Few-shot Video-to-Video Synthesis · CG characters | performance | Hurt |
Closest papers by meaning
- Video-to-Video Synthesis (NeurIPS 2018 · similarity 0.89)
- World-Consistent Video-to-Video Synthesis (ECCV 2020 · similarity 0.89)
- VLOGGER: Multimodal Diffusion for Embodied Avatar Synthesis (CVPR 2025 · similarity 0.86)
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.85)
- HumanDreamer: Generating Controllable Human-Motion Videos via Decoupled Generation (CVPR 2025 · similarity 0.85)
- BIVDiff: A Training-Free Framework for General-Purpose Video Synthesis via Bridging Image and Video Diffusion Models (CVPR 2024 · similarity 0.85)
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.
