Deepak Pathak

I pushed an update, please run this file to load the model: https://github.com/pathak22/unsupervised-video#2-fetching-motion-segmentation-models

@aditya-vora The main trick in getting it work is the Trimap loss (described in the paper). So, if you look at the validation loss on the full image without Trimap...

Actually, different versions of crf code had different way of shaping the image. I am handling it lazily in comments (i know its not clean! :( ). But yes, if...

Thanks for your interest. These lines to be changed: https://github.com/pathak22/context-encoder/blob/master/train.lua#L282-L285 https://github.com/pathak22/context-encoder/blob/master/train.lua#L418-L431 https://github.com/pathak22/context-encoder/blob/master/test.lua#L78-L85 https://github.com/pathak22/context-encoder/blob/master/test.lua#L98

I believe this is more a lua question than related to context encoders. There is this random region inpainting code here: https://github.com/pathak22/context-encoder/blob/master/train_random.lua You can take help from it, and modify...

When I was training context encoder (~2yrs ago), I found it difficult for the training to converge when the generator is an AlexNet and discriminator is same as in the...

In this code, I did not support this. But if you would like, there are several followup works which try to run this approach with high-resolution images.

@include-chenshen No particular reason as such -- just wanted to keep Discriminator, Generator, and Encoder as symmetric as possible. I do not think this choice should affect the results in...

What is a blocking artifact? Can you post some images?

Surprising! Is it the from the trained model I released? This very result looks much better when I tried. For instance see this figure here: https://people.eecs.berkeley.edu/~pathak/context_encoder/resources/result_fig.jpg Could you give more...