Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Lyu, Hengye, Li, Zisu, Hong, Yue, Weng, Yueting, Shi, Jiaxin, Zhang, Hanwang, Liang, Chen
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2604.13509
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915938022981632
author Lyu, Hengye
Li, Zisu
Hong, Yue
Weng, Yueting
Shi, Jiaxin
Zhang, Hanwang
Liang, Chen
author_facet Lyu, Hengye
Li, Zisu
Hong, Yue
Weng, Yueting
Shi, Jiaxin
Zhang, Hanwang
Liang, Chen
contents Recent advances in video generation models has significantly accelerated video generation and related downstream tasks. Among these, video stylization holds important research value in areas such as immersive applications and artistic creation, attracting widespread attention. However, existing diffusion-based video stylization methods struggle to maintain stability and consistency when processing long videos, and their high computational cost and multi-step denoising make them difficult to apply in practical scenarios. In this work, we propose RTR-DiT (DiT as Real-Time Rerenderer), a steaming video stylization framework built upon Diffusion Transformer. We first fine-tune a bidirectional teacher model on a curated video stylization dataset, supporting both text-guided and reference-guided video stylization tasks, and subsequently distill it into a few-step autoregressive model via post-training with Self Forcing and Distribution Matching Distillation. Furthermore, we propose a reference-preserving KV cache update strategy that not only enables stable and consistent processing of long videos, but also supports real-time switching between text prompts and reference images. Experimental results show that RTR-DiT outperforms existing methods in both text-guided and reference-guided video stylization tasks, in terms of quantitative metrics and visual quality, and demonstrates excellent performance in real-time long video stylization and interactive style-switching applications.
format Preprint
id arxiv_https___arxiv_org_abs_2604_13509
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DiT as Real-Time Rerenderer: Streaming Video Stylization with Autoregressive Diffusion Transformer
Lyu, Hengye
Li, Zisu
Hong, Yue
Weng, Yueting
Shi, Jiaxin
Zhang, Hanwang
Liang, Chen
Computer Vision and Pattern Recognition
Recent advances in video generation models has significantly accelerated video generation and related downstream tasks. Among these, video stylization holds important research value in areas such as immersive applications and artistic creation, attracting widespread attention. However, existing diffusion-based video stylization methods struggle to maintain stability and consistency when processing long videos, and their high computational cost and multi-step denoising make them difficult to apply in practical scenarios. In this work, we propose RTR-DiT (DiT as Real-Time Rerenderer), a steaming video stylization framework built upon Diffusion Transformer. We first fine-tune a bidirectional teacher model on a curated video stylization dataset, supporting both text-guided and reference-guided video stylization tasks, and subsequently distill it into a few-step autoregressive model via post-training with Self Forcing and Distribution Matching Distillation. Furthermore, we propose a reference-preserving KV cache update strategy that not only enables stable and consistent processing of long videos, but also supports real-time switching between text prompts and reference images. Experimental results show that RTR-DiT outperforms existing methods in both text-guided and reference-guided video stylization tasks, in terms of quantitative metrics and visual quality, and demonstrates excellent performance in real-time long video stylization and interactive style-switching applications.
title DiT as Real-Time Rerenderer: Streaming Video Stylization with Autoregressive Diffusion Transformer
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.13509