Evaluating Design Video Generation: Metrics for Compositional Fidelity
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917500869935104 |
|---|---|
| author | Deganutti, Adrienne Cao, Dingning Seol, Jaejung Hirsch, Elad Mehta, Purvanshi |
| author_facet | Deganutti, Adrienne Cao, Dingning Seol, Jaejung Hirsch, Elad Mehta, Purvanshi |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_16223 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Evaluating Design Video Generation: Metrics for Compositional Fidelity Deganutti, Adrienne Cao, Dingning Seol, Jaejung Hirsch, Elad Mehta, Purvanshi Graphics Artificial Intelligence Computer Vision and Pattern Recognition 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. |
| title | Evaluating Design Video Generation: Metrics for Compositional Fidelity |
| topic | Graphics Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.16223 |