MoFu: Scale-Aware Modulation and Fourier Fusion for Multi-Subject Video Generation
Run Ling, Ke Cao, Jian Lu, Ao Ma, Haowei Liu, Runze He, Changwei Wang, Rongtao Xu
Abstract
Multi-subject video generation aims to synthesize videos from textual prompts and multiple reference images, ensuring that each subject preserves natural scale and visual fidelity. However, current methods face two challenges: scale inconsistency, where variations in subject size lead to unnatural generation, and permutation sensitivity, where the order of reference inputs causes subject distortion. In this paper, we propose MoFu, a unified framework that tackles both challenges. For scale inconsistency, we introduce Scale-Aware Modulation (SMO), an LLM-guided module that extracts implicit scale cues from the prompt and modulates features to ensure consistent subject sizes. To address permutation sensitivity, we present a simple yet effective Fourier Fusion strategy that processes the frequency information of reference features via the Fast Fourier Transform to produce a unified representation. Besides, we design a Scale-Permutation Stability Loss to jointly encourage scale-consistent and permutation-invariant generation. To further evaluate these challenges, we establish a dedicated benchmark with controlled variations in subject scale and reference permutation. Extensive experiments demonstrate that MoFu significantly outperforms existing methods in preserving natural scale, subject fidelity, and overall visual quality.
BibTeX
@inproceedings{aaai2026_mofuscaleawaremo,
title = {MoFu: Scale-Aware Modulation and Fourier Fusion for Multi-Subject Video Generation},
author = {Run Ling and Ke Cao and Jian Lu and Ao Ma and Haowei Liu and Runze He and Changwei Wang and Rongtao Xu and Yihua Shao and Zhanjie Zhang and Peng Wu and Guibing Guo and Wei Feng and Zheng Zhang and Jingjing Lv and Junjie Shen and Ching Law and Xingwei Wang},
booktitle = {AAAI 2026},
year = {2026}
}