6 papers
CDST: Color Disentangled Style Transfer for Universal Style Reference Customization
Shiwen Zhang, Zhuowei Chen, Lang Chen +1
We introduce Color Disentangled Style Transfer (CDST), a novel and efficient two-stream style transfer training paradigm which completely isolates color from style and forces the s…
OmniTransfer: All-in-one Framework for Spatio-temporal Video Transfer
Pengze Zhang, Yanze Wu, Mengtian Li +8
Videos convey richer information than images or text, capturing both spatial and temporal dynamics. However, most existing video customization methods rely on reference images or t…
DreamO: A Unified Framework for Image Customization
Chong Mou, Yanze Wu, Wenxu Wu +15
Recently, extensive research on image customization (e.g., identity, subject, style, background, etc.) demonstrates strong customization capabilities in large-scale generative mode…
InstructX: Towards Unified Visual Editing with MLLM Guidance
Chong Mou, Qichao Sun, Yanze Wu +5
With recent advances in Multimodal Large Language Models (MLLMs) showing strong visual understanding and reasoning, interest is growing in using them to improve the editing perform…
MUSAR: Exploring Multi-Subject Customization from Single-Subject Dataset via Attention Routing
Zinan Guo, Pengze Zhang, Yanze Wu +3
Current multi-subject customization approaches encounter two critical challenges: the difficulty in acquiring diverse multi-subject training data, and attribute entanglement across…
PuLID: Pure and Lightning ID Customization via Contrastive Alignment
Zinan Guo, Yanze Wu, Zhuowei Chen +3
We propose Pure and Lightning ID customization (PuLID), a novel tuning-free ID customization method for text-to-image generation. By incorporating a Lightning T2I branch with a sta…