Today's article introduces you to one of the latest greatVirtual Try-On Solutions:catvton-fluxIt's a combination of CATVTON (CatVTON: Concatenation Is All You Need for Virtual Try-On with Diffusion Models) The powerful features andFlux repainted restoration modelin order toRealize true and accurate clothing transfers. At the same timecatvton-fluxThe solution is also influenced byAli In-Context LoRA prompts project inspiration.
- Github:: https://github.com/nftblackmagic/catvton-flux
- The catvton-flux-alpha model:: https://huggingface.co/xiaozaa/catvton-flux-alpha/tree/main
- catvton-flux-lora-alpha modeling:: https://huggingface.co/xiaozaa/catvton-flux-lora-alpha
InComfyUIPlugin Manager by means of theGit Installationto installComfyUI-CatvtonFluxWrapperPlug-ins. AlsoRequires science to automatically download the model, import the workflow to start the experience.
- ComfyUI-CatvtonFluxWrapper Plugin:: https://github.com/lujiazho/ComfyUI-CatvtonFluxWrapper
- sam2-safetensors: Download the model and place it in the catalog/ComfyUI/models/sam2 . Download: https://huggingface.co/Kijai/sam2-safetensors/blob/main/sam2_hiera_tiny.safetensors
- xiaozaa/catvton-flux-lora-alpha: The model will be downloaded and placed in the huggface cache directory by itself. Download: https://huggingface.co/xiaozaa/catvton-flux-lora-alpha/tree/main
Flux Wensheng Diagram Workflow
Students interested in Flux literate diagrams can refer to LIBLIB to run workflows online:FLUX[Continued]: 12B parameter 23G largest open source Vincennes model, Dev version straight out of the stunning beauty of the picture appreciation
The ComfyUI workflow and models mentioned in this article can be downloaded on LIBLIBAI or run online for experience:
- F.1 - Cherry Dreams and Flowing Lights - Watery Mekong Geling Fragrance:
https://www.liblib.art/modelinfo/134c6dd95aef48e98a22b24e003e026b
- Workflow - Flux text | graphs to generate graphs + LORA + prompts backpropagation one-click switch workflow:
https://www.liblib.art/modelinfo/782aacd70f604da39e83368c696a02a8
catvton-flux virtual changing workflow
The catvton-flux Virtual Dressing Workflow has been uploaded to the LIBLIB platform:
https://www.liblib.art/modelinfo/c4286818494747b284d0f0ab78d88d0b?versionUuid=5a527b0d2f084af487bd6d3d0c6571c4
Notice:
- catvton-fluxIncludes both large model and LORA implementations. The large model approach requires more than 40G of video memory. For consumer graphics memory, you need to wait for the community GGUF quantized version.
- LORA implementationIncludes two methods of subject character mask recognition:apply by handandSAM2 identificationThe reader can choose the corresponding way to test it by himself.
- Online Experience:: https://huggingface.co/spaces/xiaozaa/catvton-flux-try-on
