Nicolas Ugrinovic

Hi @Arthur151, sorry for the delay. I'll try to upload the `d_{ord}` eval function as soon as possible and release the rest after that.

@Arthur151 you now have the function to calculate depth order [here](https://github.com/nicolasugrinovic/size_depth_disambiguation/blob/62275d094ebb3086359e8b9fde0eda0ee0e29318/eval.py#L330). Basically a matrix operation to know which persons are behind or in front of the other persons.

What do you mean by `equal class`? Yes, some images where it is not possible to do an ID matching of the estimated person with its corresponding GT are not...

@Arthur151 you can check the file [done_files_eval.csv](https://github.com/nicolasugrinovic/size_depth_disambiguation/blob/main/done_files_eval.csv). It is a list of all the images used for the evaluation. The evaluation code discards some examples in which it is impossible...

Did you solve this? Are you still having that error message? I get something similar when the code is executed: > error in nnd updateOutput: no kernel image is available...

@[raypine](https://github.com/raypine) could you provide the weights trained with Panoptic dataset?

Could you provide more detail about the training time? How many GPUs did you use and how long did it take you to train? Thanks

Hi, Thanks for sharing your code on this great work. Can you provide the **pre-trained model weights** for testing in the MuPoTS-3D dataset? or can I use the weights you...

@jiahaoLjh any updates on the release of the trained weights on MuCo?

Hi kudos for the great work. Yes, I was wondering the same. Having the weights for this model would be great and will ease the comparison with your model!