新的基准和方法解决了视频生成中的视觉文本渲染和编辑问题

报道来源 [7]

1. arXiv cs.AI TIER_1 English(EN) · Xinghao Chen, Xiangbo Gao, Jiongze Yu, Yuheng Wu, Zhengzhong Tu · 2026-10-01 04:00

   ViTeX-Bench：高保真视频场景文本编辑的基准测试

   arXiv:2609.40356v1 Announce Type: cross Abstract: Recent video generation is increasingly realistic and controllable, yet video editing remains less developed, particularly for precise local edits that must preserve the original scene dynamics. Video scene text editing replaces t…

2. Hugging Face Daily Papers TIER_1 English(EN) · 2026-10-01 00:00

   VTR-Bench：用于评估视频生成中视觉文本渲染的系统性基准测试

   Recent video generation models can produce highly realistic videos from natural language instructions, with visual quality approaching cinematic standards. Existing evaluation benchmarks, however, predominantly assess visual quality, aesthetic appeal and physical plausibility, wh…

3. arXiv cs.CV TIER_1 English(EN) · Yu Huang, Jungang Li, Zhiyuan Wang, Yonghua Hei, Song Dai, Jiayu Yang, Deyuan Liu, Xiang Zheng, Xiaoshuang Shi, Hao Cheng, Kaidi Xu · 2026-10-02 04:00

   VTR-Bench：用于评估视频生成中视觉文本渲染的系统性基准测试

   arXiv:2610.01499v1 Announce Type: new Abstract: Recent video generation models can produce highly realistic videos from natural language instructions, with visual quality approaching cinematic standards. Existing evaluation benchmarks, however, predominantly assess visual quality…

4. arXiv cs.CV TIER_1 English(EN) · Fa-Ting Hong, Peter Wonka · 2026-10-02 04:00

   SuperMotion：面向文本驱动人体运动编辑的保留源的去噪方法

   arXiv:2610.01517v1 Announce Type: new Abstract: Text-driven human motion editing aims to realize a requested change while preserving compatible source content. Existing diffusion editors rely largely on learned conditioning for preservation of the unedited part, yet their outputs…

5. arXiv cs.CV TIER_1 English(EN) · Taewon Kang, Yu Shen, Ming C. Lin · 2026-10-01 04:00

   受动力学启发的扩散模型用于前景保留的文档背景编辑

   arXiv:2601.21857v2 Announce Type: replace Abstract: We revisit diffusion-based generation for structured visual content and identify a fundamental limitation of existing approaches: foreground preservation and background stylization are typically enforced through external interve…

6. arXiv cs.CV TIER_1 English(EN) · Nicolas Thiebaut, Nameer Hirschkind, Xiao Yu, Kyle Spence · 2026-10-01 04:00

   STEPS：使用扩散和对比风格编码进行风格保留的场景文本编辑

   arXiv:2609.38636v1 Announce Type: new Abstract: We introduce Scene Text Editing with Preserved Style (STEPS), a novel diffusion model architecture for quality text replacement in images. Scene Text Editing (STE), also known as Visual Text Editing, consists of changing the textual…

7. arXiv cs.CV TIER_1 English(EN) · Ziying Zhang, Litao Li, Junchao Liao, Tianyi Zeng, Siyu Zhu, Long Qin, Zhenghao Zhang · 2026-09-30 04:00

   超越可读性：统一视频生成中的视觉文本渲染与原地编辑基准测试

   arXiv:2609.36598v1 Announce Type: new Abstract: A video can exhibit convincing motion and photorealism yet fail immediately when visual text collapses. Unlike generic scene content, visual text is unforgiving in video generation: minor stroke corruption, temporal instability, or …