Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2512.22310 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917171998752768 |
|---|---|
| author | Ling, Run Cao, Ke Lu, Jian Ma, Ao Liu, Haowei He, Runze Wang, Changwei Xu, Rongtao Shao, Yihua Zhang, Zhanjie Wu, Peng Guo, Guibing Feng, Wei Zhang, Zheng Lv, Jingjing Shen, Junjie Law, Ching Wang, Xingwei |
| author_facet | Ling, Run Cao, Ke Lu, Jian Ma, Ao Liu, Haowei He, Runze Wang, Changwei Xu, Rongtao Shao, Yihua Zhang, Zhanjie Wu, Peng Guo, Guibing Feng, Wei Zhang, Zheng Lv, Jingjing Shen, Junjie Law, Ching Wang, Xingwei |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_22310 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MoFu: Scale-Aware Modulation and Fourier Fusion for Multi-Subject Video Generation Ling, Run Cao, Ke Lu, Jian Ma, Ao Liu, Haowei He, Runze Wang, Changwei Xu, Rongtao Shao, Yihua Zhang, Zhanjie Wu, Peng Guo, Guibing Feng, Wei Zhang, Zheng Lv, Jingjing Shen, Junjie Law, Ching Wang, Xingwei Computer Vision and Pattern Recognition 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. |
| title | MoFu: Scale-Aware Modulation and Fourier Fusion for Multi-Subject Video Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.22310 |