olivier-gillet

Thank you for your answer. Sure. But I want to try the power of GANs for this task. I was just surprised that theses dots are exactly (255, 0, 0)...

PatchGAN corresponds to the discriminator part : https://github.com/aitorzip/PyTorch-CycleGAN/blob/67da8f9e2b69bd68763451803c7700aaccc92f18/models.py#L64

I think 4GB of GPU ram is not enough to train all the cycle GAN models. Personally I have 11GB (1080 Ti) of GPU ram, and the max I can...

Can you provide another link for non chinese speaking people ?

It is possible to download the model trained with kerypoints at : https://github.com/Superlee506/Mask_RCNN_Humanpose/releases/download/v0.9-alpha/mask_rcnn_coco_humanpose.h5