# MoFu: Scale-Aware Modulation and Fourier Fusion for Multi-Subject Video Generation

arXiv: https://arxiv.org/abs/2512.22310 · HTML: https://mlsanity.com/paper/2512.22310/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 45 references that are in the map.
- First posted on arXiv 26 Dec 2025.

## Builds on (most cited first, 27 of 27)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [LTX-Video: Realtime Video Latent Diffusion](https://mlsanity.com/paper/2501.00103/) (2024)
- [VACE: All-in-One Video Creation and Editing](https://mlsanity.com/paper/2503.07598/) (2025)
- [FastComposer: Tuning-Free Multi-Subject Image Generation with Localized Attention](https://mlsanity.com/paper/2305.10431/) (2023)
- [MAGI-1: Autoregressive Video Generation at Scale](https://mlsanity.com/paper/2505.13211/) (2025)
- [Identity-Preserving Text-to-Video Generation by Frequency Decomposition](https://mlsanity.com/paper/2411.17440/) (2024)
- [Phantom: Subject-consistent video generation via cross-modal alignment](https://mlsanity.com/paper/2502.11079/) (2025)
- [WISA: World Simulator Assistant for Physics-Aware Text-to-Video Generation](https://mlsanity.com/paper/2503.08153/) (2025)
- [CustomVideo: Customizing Text-to-Video Generation with Multiple Subjects](https://mlsanity.com/paper/2401.09962/) (2024)
- [SkyReels-A2: Compose Anything in Video Diffusion Transformers](https://mlsanity.com/paper/2504.02436/) (2025)
- [OpenS2V-Nexus: A Detailed Benchmark and Million-Scale Dataset for Subject-to-Video Generation](https://mlsanity.com/paper/2505.20292/) (2025)
- [ConceptMaster: Multi-Concept Video Customization on Diffusion Transformer Models Without Test-Time Tuning](https://mlsanity.com/paper/2501.04698/) (2025)
- [Multi-subject Open-set Personalization in Video Generation](https://mlsanity.com/paper/2501.06187/) (2025)
- [ArtBank: Artistic Style Transfer with Pre-trained Diffusion Model and Implicit Style Prompt Bank](https://mlsanity.com/paper/2312.06135/) (2023)
- [Movie Weaver: Tuning-Free Multi-Concept Video Personalization with Anchored Prompts](https://mlsanity.com/paper/2502.07802/) (2025)
- [DropoutGS: Dropping Out Gaussians for Better Sparse-view Rendering](https://mlsanity.com/paper/2504.09491/) (2025)
- [RelaCtrl: Relevance-Guided Efficient Control for Diffusion Transformers](https://mlsanity.com/paper/2502.14377/) (2025)
- [FancyVideo: Towards Dynamic and Consistent Video Generation via Cross-frame Textual Guidance](https://mlsanity.com/paper/2408.08189/) (2024)
- [Learning Robust Stereo Matching in the Wild with Selective Mixture-of-Experts](https://mlsanity.com/paper/2507.04631/) (2025)
- [Uni-Layout: Integrating Human Feedback in Unified Layout Generation and Evaluation](https://mlsanity.com/paper/2508.02374/) (2025)
- [Tora2: Motion and Appearance Customized Diffusion Transformer for Multi-Entity Video Generation](https://mlsanity.com/paper/2507.05963/) (2025)
- [Qihoo-T2X: An Efficient Proxy-Tokenized Diffusion Transformer for Text-to-Any-Task](https://mlsanity.com/paper/2409.04005/) (2024)
- [Lay2Story: Extending Diffusion Transformers for Layout-Togglable Story Generation](https://mlsanity.com/paper/2508.08949/) (2025)
- [StyleDiT: A Unified Framework for Diverse Child and Partner Faces Synthesis with Style Latent Diffusion Transformer](https://mlsanity.com/paper/2412.10785/) (2024)

## Built on by (most cited first, 6 of 6)

- [OmniShow: Unifying Multimodal Conditions for Human-Object Interaction Video Generation](https://mlsanity.com/paper/2604.11804/) (2026)
- [InnoAds-Composer: Efficient Condition Composition for E-Commerce Poster Generation](https://mlsanity.com/paper/2603.05898/) (2026)
- [HiFi-Inpaint: Towards High-Fidelity Reference-Based Inpainting for Generating Detail-Preserving Human-Product Images](https://mlsanity.com/paper/2603.02210/) (2026)
- [InnoText: A Unified Model for Visual Text Generation and Editing](https://mlsanity.com/paper/2607.22101/) (2026)
- [HOMIE: Human-object Centric Video Personalization via Multimodal Intelligent Enhancement](https://mlsanity.com/paper/2607.18217/) (2026)
- [FedCVU: Federated Learning for Cross-View Video Understanding](https://mlsanity.com/paper/2603.21647/) (2026)

## Abstract (arXiv)

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.