Benchmark RadarAnimation2Code
Animation2Code evaluates temporal visual reasoning in video-to-code generation. It includes 1,069 web animation videos with corresponding HTML/CSS/JavaScript implementations, and uses appearance and temporal similarity metrics to assess model performance.
- Released
- 2026-06-26
- Readiness
- Inspectable
- Primary field
- General AI
Why it matters
Animation2Code addresses the lack of benchmarks for temporal dynamics in visual-to-code tasks. It provides a way to measure both visual fidelity and temporal alignment, which is crucial for applications requiring precise animation reconstruction.
Motivation
While recent vision-language models (VLMs) have achieved significant improvements on static visual-to-code tasks such as generating code for webpages, charts, or SVGs, it remains unclear whether they can recover temporal dynamics when motion is present.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.