pytorch_Realtime_Multi-Person_Pose_Estimation icon indicating copy to clipboard operation
pytorch_Realtime_Multi-Person_Pose_Estimation copied to clipboard

Results 61 pytorch_Realtime_Multi-Person_Pose_Estimation issues
Sort by recently updated
recently updated
newest added

First of all, thank you for the great work! I think there might be a bug in the add_neck function: i think the neck has to be placed only if...

Hello, I thing I run the right way your code on my computer (with the last version of pytorch) but I did not get the same results. With VGG19, I...

I've noticed that in addition to original VGG version, you also implemented a Stacked Hourglass version. Is Stacked Hourglass version any better? The result you report on readme.md is from...

Hi, sorry silly question, I am a newbie to COCO and openPose, I am just wondering where should I put all the datasets/mask/json and what directory should I pass to...

Hi, In coco dataloader of training stage, you generate heatmap by sigma=7.0 regardless of its scale. And in evaluation, you use multi-scale images as input. To my understand, heatmap is...

windows10+RtX2070+CUDA9.0+torch1.0.1+python3.6.5,this is my enviroment.When I run picture_demo.py,the error of "RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED" appears.why?Have you encountered a similar problem? Do you have a solution?

https://github.com/tensorboy/pytorch_Realtime_Multi-Person_Pose_Estimation/blob/b0013fcf232896b085101ec45e41bca220bf655e/network/post.py#L298 Hi, this is probably too detailed and yet i think this is a potential bug - why not consider the `if person_limbs[joint_src_type] != limb_info[0]` case?

Hi, It seems that the length constraint is also adopted from the limb_width (https://github.com/tensorboy/pytorch_Realtime_Multi-Person_Pose_Estimation/blob/master/training/datasets/coco_data/paf.py#L40). Is that so?

can you tell me the ultimate loss(train/val) in your training?:)