Evaluating Design Video Generation: Metrics for Compositional Fidelity
图形学
2026-05-18 v1 人工智能
计算机视觉与模式识别
摘要
Generative video models are increasingly used in design animation tasks, yet no standardized evaluation framework exists for this domain. Unlike natural video generation, design animation imposes structured constraints: specific components shall animate with prescribed motion types, directions, speed and timing, while non-animated regions must remain stable and layout structure must be preserved. This paper provides a fully automated evaluation framework organized across four dimensions: layout fidelity, motion correctness, temporal quality, and content fidelity. This eliminates the reliance on subjective human evaluation and establishes a common basis for benchmarking progress in the field.
引用
@article{arxiv.2605.16223,
title = {Evaluating Design Video Generation: Metrics for Compositional Fidelity},
author = {Adrienne Deganutti and Dingning Cao and Jaejung Seol and Elad Hirsch and Purvanshi Mehta},
journal= {arXiv preprint arXiv:2605.16223},
year = {2026}
}