# What is it?

Multi-view diffusion is a type of diffusion model (e.g. [Stable Diffusion](https://huggingface.co/blog/stable_diffusion)). However, instead of being trained on regular images, it's trained on multiple views of an object from different perspectives.

![Multi-view diffusion](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/ml-for-3d-course/multiview.webp)

## Problems

Multi-view diffusion is useful for 3D.

However, it doesn't work out-of-the-box. It tends to suffer from something called the Janus problem, where objects have multiple faces (or, more generally, lack of consistency across views).

![Janus problem](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/ml-for-3d-course/janus.png)

## Solutions

State-of-the-art multi-view-diffusion models like [MVDream](https://huggingface.co/MVDream/MVDream) address this problem using specialized techniques.

I won't be going into technical detail on multi-view diffusion in this course, since it's more related to diffusion than to 3D, but if you'd like to learn more, check out the [Diffusion course](https://huggingface.co/learn/diffusion-course/unit0/1).

In the next sections, we'll be using a pre-trained multi-view diffusion model to generate multi-view images.

