Saved in:
Bibliographic Details
Main Authors: Wang, Zixuan, Zhou, Ziqin, Chen, Feng, Peng, Duo, Hu, Yixin, Li, Changsheng, Lei, Yinjie
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.09104
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914431226609664
author Wang, Zixuan
Zhou, Ziqin
Chen, Feng
Peng, Duo
Hu, Yixin
Li, Changsheng
Lei, Yinjie
author_facet Wang, Zixuan
Zhou, Ziqin
Chen, Feng
Peng, Duo
Hu, Yixin
Li, Changsheng
Lei, Yinjie
contents Compositional video generation aims to synthesize multiple instances with diverse appearance and motion. However, current approaches mainly focus on binding semantics, neglecting to understand diverse motion categories specified in prompts. In this paper, we propose a motion factorization framework that decomposes complex motion into three primary categories: motionlessness, rigid motion, and non-rigid motion. Specifically, our framework follows a planning before generation paradigm. (1) During planning, we reason about motion laws on the motion graph to obtain frame-wise changes in the shape and position of each instance. This alleviates semantic ambiguities in the user prompt by organizing it into a structured representation of instances and their interactions. (2) During generation, we modulate the synthesis of distinct motion categories in a disentangled manner. Conditioned on the motion cues, guidance branches stabilize appearance in motionless regions, preserve rigid-body geometry, and regularize local non-rigid deformations. Crucially, our two modules are model-agnostic, which can be seamlessly incorporated into various diffusion model architectures. Extensive experiments demonstrate that our framework achieves impressive performance in motion synthesis on real-world benchmarks. Code is available at https://github.com/ZixuanWang0525/MF-CVG.
format Preprint
id arxiv_https___arxiv_org_abs_2603_09104
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Training-free Motion Factorization for Compositional Video Generation
Wang, Zixuan
Zhou, Ziqin
Chen, Feng
Peng, Duo
Hu, Yixin
Li, Changsheng
Lei, Yinjie
Computer Vision and Pattern Recognition
Compositional video generation aims to synthesize multiple instances with diverse appearance and motion. However, current approaches mainly focus on binding semantics, neglecting to understand diverse motion categories specified in prompts. In this paper, we propose a motion factorization framework that decomposes complex motion into three primary categories: motionlessness, rigid motion, and non-rigid motion. Specifically, our framework follows a planning before generation paradigm. (1) During planning, we reason about motion laws on the motion graph to obtain frame-wise changes in the shape and position of each instance. This alleviates semantic ambiguities in the user prompt by organizing it into a structured representation of instances and their interactions. (2) During generation, we modulate the synthesis of distinct motion categories in a disentangled manner. Conditioned on the motion cues, guidance branches stabilize appearance in motionless regions, preserve rigid-body geometry, and regularize local non-rigid deformations. Crucially, our two modules are model-agnostic, which can be seamlessly incorporated into various diffusion model architectures. Extensive experiments demonstrate that our framework achieves impressive performance in motion synthesis on real-world benchmarks. Code is available at https://github.com/ZixuanWang0525/MF-CVG.
title Training-free Motion Factorization for Compositional Video Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.09104