Grégoire Payen de La Garanderie

I am not able to reproduce your problem. I highly suspect that the problem might be due to your older version of GCC. Could you try with a more recent...

Hi @ShravanthiPatil , You are welcome. I am not currently actively using the Waymo dataset and I haven’t tried yet the updated dataset so I might be of limited help....

Thanks for reporting the issue. I am not currently actively using the dataset, so I am not able to test the proposed fix; however, your observation sounds correct. Feel free...

This is to transform to image space. According to the Waymo data description: “The origin of this frame is set to the vehicle position when the vehicle starts. It is...

Thanks for the figure. If you look at the camera frame shown in the figure, it’s front,left,up. I think that it’s a little bit unusual to choose this space for...

In the original dataset, there were labels for objects that were either very heavily occluded or completely hidden (for instance behind a wall). The filtering of objects with less than...

Hi, I am afraid not out of the box. It might be possible but you would have to port some of the operations from the GPU to CPU.

I’ll try: Given a list of bounding boxes [a,b,c,d,e,f] with a,b,e overlapping each other and c,f are also overlapping each other and d overlapping none. Assuming that the output of...

Not always. You have the guarantee that `top_k >= num_to_keep`. num_to_keep can be smaller if either the size of your input array is smaller than top_k or your boxes are...

Hello, Sorry, I forgot to push the script to Github; I have just pushed it now at: https://github.com/gdlg/pytorch-CycleGAN-and-pix2pix The two scripts are essentially doing the same thing. The test_single.py will...