ibaigorordo / onnx-mobile-human-pose-3d Goto Github PK

I can't download through the link（https://github.com/PINTO0309/PINTO_model_zoo/blob/main/135_CoEx/download.sh）.The error screenshot is as follows：

，I can not find the cookie file.Can you provide an absolute download link，thank you.

Hi,thanks for the great work. I follow your Readme and download the required models including Mobile Human Pose and YOLOv5s, and below is the inference result and which is not as accurate as i expect.
The input image size is 512*512, and i have no idea to choose the proper detection model(different resolution),so i just tried it all,which not improved the result.Could you point out what the problem is?

Hi, could you plz upload the original pytorch model? Thx.

I have tested on rtsp live feed to get 2D and 3D poses. but i see so much variation on keypoints extracted due to which even 2D visualization is not looking good. How can we make it better, is there any better model file available. and what is role of Focal length and principal points in computation because i see they are commented in pix2cam function.

Thank you for sharing.
Could you tell me the idea behind the following line?
https://github.com/ibaiGorordo/ONNX-Mobile-Human-Pose-3D/blob/main/mobileHumanPose/mobileHumanPose.py#L161

I understand you need abs_depth. For instance, using the average height will be one of solutions.
focal_length(fixed value) : abs_depth(unknown value) = height_in_pixel(known value) : average_height(fixed value)
Using the above equation, we can get abs_depth. (It may not work well, though)

In your code, it looks you use area, and add +500. The values may be determined by experiments, but could you tell me the basic idea of this?