Håkon Hukkelås

Hi! I'll take a look at this today. Due to the messy license of the images in the COCO dataset, I will only release the annotations from CSE. Along with...

The [docs/DATASETS.md](https://github.com/hukkelas/full_body_anonymization/blob/main/docs/DATASETS.md) is now updated. Please notify me if you have any issues with downloading the COCO-cse dataset.

Thanks for notifying me! I fixed the typos now :) Regarding the error in train.py, it's due to a bug in the coco-cse downloading script. [This commit](https://github.com/hukkelas/full_body_anonymization/commit/c61745b137c84ffb742ef6ab2f4721db4acf22b7) fixes the issue....

Can you give a full stacktrace of your error?

Code for generating embeddings for new training datasets is currently not supported, but it should not be that much required. The following code detects an image, crops the different detections...

My pleasure :)

Model download is done with torch.hub. You can change the save directory by setting the environment variable `TORCH_HOME`. If you run ``` TORCH_HOME=~/Downloads python3 test.py ``` the model will be...

Could you give a bit more information to reproduce this? E.g. your python/pytorch/numpy versions and operating system. Also, please provide the exact script you ran to reproduce this.

Hi, I don't have access to a GPU currently as I'm on vacation. Will be able to inspect the issue when I'm back. If anyone finds the issue, please submit...

Hi, we did not evaluate the number of FLOPs for the full network architecture.