make a video pytorch wip

We have hosted the application make a video pytorch wip in order to run this application in our online workstations with Wine or directly.

Quick description about make a video pytorch wip:

Implementation of Make-A-Video, new SOTA text to Video generator from Meta AI, in Pytorch. They combine pseudo-3d convolutions (axial convolutions) and temporal attention and show much better temporal fusion. The pseudo-3d convolutions isn't a new concept. It has been explored before in other contexts, say for protein contact prediction as "dimensional hybrid residual networks". The gist of the paper comes down to, take a SOTA text-to-image model (here they use DALL-E2, but the same learning points would easily apply to Imagen), make a few minor modifications for attention across time and other ways to skimp on the compute cost, do frame interpolation correctly, get a great Video model out. Passing in images (if one were to pretrain on images first), both temporal convolution and attention will be automatically skipped. In other words, you can use this straightforwardly in your 2d Unet and then port it over to a 3d Unet once that phase of the training is done.

Features:

• The temporal modules are initialized to output identity as the paper had done
• You can also control the two modules so that when fed 3-dimensional features, it only does training spatially
• Full SpaceTimeUnet that is agnostic to images or Video training, and where even if Video is passed in, time can be ignored
• Passing in images (if one were to pretrain on images first), both temporal convolution and attention will be automatically skipped
• The gist of the paper comes down to, take a SOTA text-to-image model
• Implementation of Make-A-Video, new SOTA text to Video generator from Meta AI, in Pytorch

Programming Language: Python.
Categories: