Moore-AnimateAnyone
Moore-AnimateAnyone copied to clipboard
3D Computation Clarity in unet_3d, resnet_3d, and transform_3d
I have a question: unet_3d, resnet_3d, and transform_3d only deal with dimensional transformations, yet there is no indication anywhere that 3D computation is necessary.
In stage2, motion module will be inserted into the 3d unet. This will perform temporal self-attention, which differs from 2d unet. So we keep them seperately.