FilmBench: A Film-Grade Benchmark for Cinematic Video Generation

Published
Source
arXiv
Paper number
729
Field
Computer Vision
arXiv ID
2607.24241

Key points

  • It constructed 1,169 prompts reverse-engineered by professional directors from scenes in award-winning films, of which 1,056 are multi-shot.
  • It evaluates cinematographic language along 3 axes, 12 elements, 35 detailed metrics for T2V, and 3 metrics specific to R2V.
  • The FilmOps automatic evaluator reproduced human rankings of 9 T2V and 7 R2V models with Spearman correlations of 0.95 and 0.96, respectively.
  • It enables comparison of dynamic aesthetics and multi-shot continuity, which existing web-video-centered evaluations miss, against filmmaking standards.
  • However, because the benchmark is tailored to film and professional video production, it does not directly represent performance on short vertical videos or general user-generated videos.

Paper links

External research summaries. These are not HDATF publications or measured product results.

Read original (opens in a new tab)