ubisoft/ubisoft-laforge-animation-dataset resource

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1568 stars · 327 forks observed · 2026-08-28

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

The Ubisoft La Forge Animation Dataset (LAFAN1) is a motion capture dataset of 5 subjects, 77 sequences, and ~4.6 hours of character animation data in BVH format, released alongside code for the SIGGRAPH 2020 'Robust Motion In-betweening' paper. It includes benchmarking code for evaluating motion in-betweening models and is licensed under CC BY-NC-ND 4.0.

Use cases

• train deep learning models for character motion in-betweening
• benchmark motion transition generation algorithms
• research data-driven character animation and motion matching
• train physics-based character controllers from mocap data
• denoise or retarget optical motion capture data
• study locomotion and dance motion sequences for animation research

When to choose

• you need a large, high-quality mocap dataset for character animation research
• you want to reproduce or compare against published motion in-betweening results
• you need BVH-format motion data for motion matching or learned motion synthesis
• you are doing academic research on data-driven character animation

When to avoid

• you need a permissively licensed dataset for commercial products (CC BY-NC-ND forbids commercial use and derivatives)
• you need real-time animation tooling rather than raw data and research code
• you need mocap of objects, faces, or animals rather than human characters
• you cannot handle large files via git LFS

Facets

dataset · maturity stable

machine-learning animation benchmarking data-generation machine-learning computer-vision graphics cross-platform python motion-capture character-animation bvh motion-in-betweening deep-learning siggraph game-development research

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ubisoft/ubisoft-laforge-animation-dataset")