Haofei Xu

For stereo matching, we should modify the cross-attention to 1D because the cross-attention models cross-view interactions via cross-view similarities. Thus it's redudant to perform 2D cross-attention because the corresponding pixels...

> So if I have understood correctly, in TransformerBlock we run first self-attention, then cross-attention and lastly ffn. Both self-attention and cross-attention calls TranformerLayer class but the difference is that...

> Small update: I trained the network for stereo matching task with sceneflow and it is producing some weird looking checkerboard artifact. > >  > > Do you have...

Hi all, sorry for the late response. If this issue is still relavant to you, I would suggest to try our new GMStereo model: https://haofeixu.github.io/unimatch/ & https://github.com/autonomousvision/unimatch. No CUDA op...

The results in Table 1 is obtained on KITTI training set. You can find this information from our paper. 

Hi, I haven't tested with CUDA11. I would recommend you to have a try and see what happens.

Have you successfully compiled the deform_conv package?

Thanks @zyl1336110861 for sharing your solution! Hope it can be helpful for others!

Hi all, sorry for the late response. If this issue is still relavant to you, I would suggest to try our new GMStereo model: https://haofeixu.github.io/unimatch/ & https://github.com/autonomousvision/unimatch. No CUDA op...

Hi all, sorry for the late response. If this issue is still relavant to you, I would suggest to try our new GMStereo model: https://haofeixu.github.io/unimatch/ & https://github.com/autonomousvision/unimatch. No CUDA op...