NVlabs/few-shot-vid2vid

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1798 stars · 274 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A PyTorch implementation of few-shot photorealistic video-to-video translation from NVIDIA Research (NeurIPS 2019), generating realistic videos from poses, edge maps, or semantic label maps. The repository is deprecated in favor of the NVlabs/imaginaire repo.

Use cases

• generate human motion videos from pose sequences
• synthesize talking head videos from edge maps
• turn semantic segmentation label maps into photorealistic videos
• reproduce few-shot vid2vid research results
• train video-to-video translation models on custom datasets

When to choose

• you need the original reference implementation of the NeurIPS 2019 few-shot vid2vid paper
• you are doing academic research on video-to-video synthesis
• you want to study the model architecture before porting it

When to avoid

• you want maintained code or new features - use NVlabs/imaginaire instead
• you need a permissive license for commercial use - it is restricted to academic research
• you lack an NVIDIA GPU with CUDA
• you need production-ready video synthesis tooling

Facets

library · maturity abandoned

machine-learning deep-learning image-processing video-processing computer-vision deep-learning machine-learning python pytorch video-to-video-translation gan few-shot-learning research-code deprecated nvidia pose-to-video face-synthesis video linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="NVlabs/few-shot-vid2vid")