Saved in:
Bibliographic Details
Main Authors: Wang, Lizhen, Zhu, Yongming, Ge, Zhipeng, Zheng, Youwei, Zhang, Longhao, Hu, Tianshu, Qin, Shiyang, Luo, Mingshuang, Zhang, Jiaxu, Chen, Xin, Wang, Yulong, Zheng, Zerong, Jiang, Jianwen, Liang, Chao, Chen, Weifeng, Wang, Xing, Zhang, Yuan, Gao, Mingyuan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.10103
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914256240246784
author Wang, Lizhen
Zhu, Yongming
Ge, Zhipeng
Zheng, Youwei
Zhang, Longhao
Hu, Tianshu
Qin, Shiyang
Luo, Mingshuang
Zhang, Jiaxu
Chen, Xin
Wang, Yulong
Zheng, Zerong
Jiang, Jianwen
Liang, Chao
Chen, Weifeng
Wang, Xing
Zhang, Yuan
Gao, Mingyuan
author_facet Wang, Lizhen
Zhu, Yongming
Ge, Zhipeng
Zheng, Youwei
Zhang, Longhao
Hu, Tianshu
Qin, Shiyang
Luo, Mingshuang
Zhang, Jiaxu
Chen, Xin
Wang, Yulong
Zheng, Zerong
Jiang, Jianwen
Liang, Chao
Chen, Weifeng
Wang, Xing
Zhang, Yuan
Gao, Mingyuan
contents Interactive humanoid video generation aims to synthesize lifelike visual agents that can engage with humans through continuous and responsive video. Despite recent advances in video synthesis, existing methods often grapple with the trade-off between high-fidelity synthesis and real-time interaction requirements. In this paper, we propose FlowAct-R1, a framework specifically designed for real-time interactive humanoid video generation. Built upon a MMDiT architecture, FlowAct-R1 enables the streaming synthesis of video with arbitrary durations while maintaining low-latency responsiveness. We introduce a chunkwise diffusion forcing strategy, complemented by a novel self-forcing variant, to alleviate error accumulation and ensure long-term temporal consistency during continuous interaction. By leveraging efficient distillation and system-level optimizations, our framework achieves a stable 25fps at 480p resolution with a time-to-first-frame (TTFF) of only around 1.5 seconds. The proposed method provides holistic and fine-grained full-body control, enabling the agent to transition naturally between diverse behavioral states in interactive scenarios. Experimental results demonstrate that FlowAct-R1 achieves exceptional behavioral vividness and perceptual realism, while maintaining robust generalization across diverse character styles.
format Preprint
id arxiv_https___arxiv_org_abs_2601_10103
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FlowAct-R1: Towards Interactive Humanoid Video Generation
Wang, Lizhen
Zhu, Yongming
Ge, Zhipeng
Zheng, Youwei
Zhang, Longhao
Hu, Tianshu
Qin, Shiyang
Luo, Mingshuang
Zhang, Jiaxu
Chen, Xin
Wang, Yulong
Zheng, Zerong
Jiang, Jianwen
Liang, Chao
Chen, Weifeng
Wang, Xing
Zhang, Yuan
Gao, Mingyuan
Computer Vision and Pattern Recognition
Artificial Intelligence
Interactive humanoid video generation aims to synthesize lifelike visual agents that can engage with humans through continuous and responsive video. Despite recent advances in video synthesis, existing methods often grapple with the trade-off between high-fidelity synthesis and real-time interaction requirements. In this paper, we propose FlowAct-R1, a framework specifically designed for real-time interactive humanoid video generation. Built upon a MMDiT architecture, FlowAct-R1 enables the streaming synthesis of video with arbitrary durations while maintaining low-latency responsiveness. We introduce a chunkwise diffusion forcing strategy, complemented by a novel self-forcing variant, to alleviate error accumulation and ensure long-term temporal consistency during continuous interaction. By leveraging efficient distillation and system-level optimizations, our framework achieves a stable 25fps at 480p resolution with a time-to-first-frame (TTFF) of only around 1.5 seconds. The proposed method provides holistic and fine-grained full-body control, enabling the agent to transition naturally between diverse behavioral states in interactive scenarios. Experimental results demonstrate that FlowAct-R1 achieves exceptional behavioral vividness and perceptual realism, while maintaining robust generalization across diverse character styles.
title FlowAct-R1: Towards Interactive Humanoid Video Generation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2601.10103