Jia Xue

Hi @williamrodz, I filled the left code, but the result is not good for the mpii images ``` from __future__ import absolute_import from __future__ import division from __future__ import print_function...

@junjieAI I just filled up the missed part of Qichao, please read the whole discussion. Actually I find the result is not good. So I just followed the author's implementation,...

I didn't work on this for a while, just briefly looked through the code, I guess that might because model loading error? Just recommend you go through the authors tutorial...

@ycszen I run your R18.speed and R18 with 4 1080Ti. the accuracies for last epoch models are 74.2 and 75.2. which is a little bit less than your reported accuracy...

please look at torch official website https://github.com/torch/nn/blob/master/doc/simple.md#nn.PixelShuffle

currently I don't have the pre-trained model, I will refactor and update the code in the near future. If you need the dataloader for MINC, you can look for my...

Hi, This repo is outdated, you can find the new code and model weights here https://github.com/jiaxue1993/pytorch-material-classification

I am extending the paper to a journal, I cleaned the code and it will come out soon

Hi, We generate the GTOS-mobile dataset by directly downsampling the full resolution images (2046 x 2040) into 256 x 256. If you want to get the full resolution images or...

It occurs because the pixel loss exclusively applies to the unmasked area, while the masked region employs perceptual loss and adversarial loss, emphasizing perceptual similarity. To ensure consistency in low-frequency...