MVGamba: Unify 3D Content Generation as State Space Sequence Modeling

Neural Information Processing Systems 

We address the challenge of crafting 3D content from a single image, sparse-view images, or text input, which can facilitate a broad range of applications, e.g ., Virtual Reality, immersive filming, digital gaming and animation.