#artificial intelligence
Jun 2026
SteerVTE: Seamless Video Text Editing with Style and Glyph Control
SteerVTE is introduced, a unified framework that steers a frozen video diffusion model to perform precise Video Text Editing through style and glyph control and substantially outperforms existing video editing baselines across text accuracy, style consistency, and temporal coherence.
Kai Zeng, Mo-Ran Li, Zheng-Wei Wang et al.
· arXiv.org · 5 citations