Preprint
Oct 2026
TextVBench: Evaluating Text Rendering and Preservation in Video Generation
Modern video generators can synthesize convincing scenes while rendering a few characters incorrectly, omitting requested text, or changing it across frames. Measuring these failures is harder than applying OCR independently to each frame: when a scene contains multiple text targets and multiple noisy OCR regions, the...
Rui-Qiang Zhang, Heng-Yi Wang, Chang Liu et al.
· 0 citations