CustomCrafter: Customized Video Generation with Preserving Motion and Concept Composition Abilities
Top Cited Papers
Open Access
- 11 April 2025
- journal article
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (8) , 8469-8477
Abstract
Customized video generation aims to generate high-quality videos guided by text prompts and subject's reference images. However, since it is only trained on static images, the fine-tuning process of subject learning disrupts abilities of video diffusion models (VDMs) to combine concepts and generate motions. To restore these abilities, some methods use additional video similar to the prompt to fine-tune or guide the model. This requires frequent changes of guiding videos and even re-tuning of the model when generating different motions, which is very inconvenient for users. In this paper, we propose CustomCrafter, a novel framework that preserves the model's motion generation and conceptual combination abilities without additional video and fine-tuning to recovery. For preserving conceptual combination ability, we design a plug-and-play module to update few parameters in VDMs, enhancing the model's ability to capture the appearance details and the ability of concept combinations for new subjects. For motion generation, we observed that VDMs tend to restore the motion of video in the early stage of denoising, while focusing on the recovery of subject details in the later stage. Therefore, we propose Dynamic Weighted Video Sampling Strategy. Using the pluggability of our subject learning modules, we reduce the impact of this module on motion generation in the early stage of denoising, preserving the ability to generate motion of VDMs. In the later stage of denoising, we restore this module to repair the appearance details of the specified subject, thereby ensuring the fidelity of the subject's appearance. Experimental results show that our method has a significant improvement compared to previous methods.Keywords
This publication has 11 references indexed in Scilit:
- CustomVideo: Customizing Text-to-Video Generation With Multiple SubjectsIEEE Transactions on Multimedia, 2026
- Swap Attention in Spatiotemporal Diffusions for Text-to-Video GenerationInternational Journal of Computer Vision, 2025
- LaVie: High-Quality Video Generation with Cascaded Latent Diffusion ModelsInternational Journal of Computer Vision, 2024
- RCS-Prompt: Learning Prompt to Rearrange Class Space for Prompt-Based Continual Learning2024
- Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing2024
- VideoBooth: Diffusion-based Video Generation with Image Prompts2024
- A Survey of Multimodal Controllable Diffusion ModelsJournal of Computer Science and Technology, 2024
- Self-Paced Multi-Grained Cross-Modal Interaction Modeling for Referring Expression ComprehensionIEEE Transactions on Image Processing, 2023
- MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing2023
- ELITE: Encoding Visual Concepts into Textual Embeddings for Customized Text-to-Image Generation2023