Deqing Sun

Hi Ferran, Thank you for your interest in our work. The learning rate in Figure 4 has been used to obtain the results on Sintel reported in our CVPR'18 paper...

Hi Phil, Thank you so much for sharing your findings. They are very interesting and I will look into them. Meanwhile, I will keep this issue open so that the...

It seems that you are doing unsupervised training, i.e., photometric loss. The loss described in the paper is for supervised setting. To compute photometric loss, you need to scale the...

Sorry for the confusion and late reply. ./model/pwc_net.caffemodel PWC-Net_ROB fine-tuned using Sintel, KITTI, and HD1K data; tends to perform robustly across different datasets ./model/pwc_net_chairs_things.caffemodel PWC-Net-ROB pre-trained using FlyingChairs and FlyingThings,...

Please check this example for the difference between forward and backward warping https://www.researchgate.net/figure/Forward-and-backward-image-warping-In-the-case-of-foward-warping-A-holes-can-occur_fig2_267946997

Sorry for the very late reply. It is a good observation and a great question. My thought was that the correlation/cost volume might be sufficient for this level, but I...

Can you try the installation script using Anaconda to set up the environment (copied from readme under PyTorch)? # setup environment conda create -n pwcnet_test python=2.7 anaconda conda activate pwcnet_test...

Could you please go to https://github.com/google-research/opticalflow-autoflow?