Guangming Zhu

@sumeetssaurav The code works well on my workstation.

@travisCxy Sorry, I did not receive the email about the issue. The labels of the test datasets were not released when I released this git, so I assigned 0 to...

You can obtain the RGB/Depth/Flow images of IsoGD from the link: https://pan.baidu.com/s/12qC5sVhQAPizpsr0axVehw Password: fymm

@emily19941102 The video2image was only tested on my workstation. The source code of brox is the main.cpp. You can compile it yourself. OpenCV is required. You can also use a...

The original Tensorlayer does not support the convolutional LSTM, so the tensorlayer/layers.py needs to be replaced with our tensorlayer-layers.py. Just follow the README exactly, I believe it will be OK.

@mengyaaa The seq_len should be even.

@emily19941102 What is the problem exactly? It seems that the training of this epoch has been finished, but problems occur during the process of testing or saving models.

@emily19941102 You may double-check the path of the test subset.

@peach2wanzi You can print the shape of the input before feeding it into the Input Layer. I guess that something may be wrong when you read images for the input...

Linux下使用ffmpeg指令