SteerVTE: Seamless Video Text Editing with Style and Glyph Control
2026-06-22 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors developed SteerVTE, a system to edit text in videos while keeping the style and motion consistent across frames. They built it on a pre-trained video model and added tools to understand both the style of the original text and the details of the new text at line and character levels. To improve text quality, they introduced special training methods and created a large dataset of video text samples. Their experiments show SteerVTE does better than existing methods in accurately changing text while looking realistic and smooth over time.
video diffusion modeltext editingstyle encoderglyph encodingtemporal coherencespatial-focal lossprogressive trainingvideo text datasetfrozen modelvisual realism
Authors
Kai Zeng, Moran Li, Zhengwei Wang, Yingchen Yu, Yiheng Lin, Ruichuan An, Ming Lu, Qi She, Wentao Zhang
Abstract
Visual text editing aims to precisely modify text in images and videos while preserving stylistic consistency and visual realism. Despite significant advances in the image domain, video text editing remains largely unexplored: it is a localized task demanding stroke-level precision within small text regions, which compounds the challenges of cross-frame accuracy, temporal coherence, and stylistic fidelity. We introduce SteerVTE, a unified framework that \underline{\textbf{steer}}s a frozen video diffusion model to perform precise \underline{\textbf{V}}ideo \underline{\textbf{T}}ext \underline{\textbf{E}}diting through style and glyph control. Built on a frozen diffusion transformer, SteerVTE attaches a lightweight text context adapter with two complementary modules: a style encoder capturing the original text's visual attributes, and dual-granularity glyph encoders encoding the target text at both the line and character levels. To overcome the inherently weak text rendering priors of video foundation models, we further propose a glyph-aware spatial-focal loss and a three-stage progressive training curriculum that scales from image to video data. To support large-scale training, we also develop an automatic synthesis pipeline and construct SteerVTE-1M, a dataset of one million triplets spanning diverse scenes, fonts, and stylistic effects. Extensive experiments demonstrate that SteerVTE substantially outperforms existing video editing baselines across text accuracy, style consistency, and temporal coherence.