Support MV-Adapter for multi-view generation

[huanngzh]

Model/Pipeline/Scheduler description

MV-Adapter is a creative productivity tool that seamlessly transfer text-to-image models to multi-view generators.

Highlights:

• generate 768x768 multi-view images
• work well with personalized models, LCM, ControlNet
• support text or image to multi-view (reconstruct 3D thereafter), or with geometry guidance for 3D texture generation
• arbitrary view generation

Btw, I can help implement it in diffusers. But I am not sure whether it should be implemented in the diffusers kernel like t2i_adapter and ip_adapter, or in community pipelines.

Open source status

• The model implementation is available.
• The model weights are available (Only relevant if addition is not a scheduler).

Provide useful links for the implementation

Official implementation: https://github.com/huanngzh/MV-Adapter
Model weights: https://huggingface.co/huanngzh/mv-adapter

[hlky]

whether it should be implemented in the diffusers kernel like t2i_adapter and ip_adapter, or in community pipelines.

cc @yiyixuxu