I’m following along with the pre-trained code for AlphaPose here:
I’ve been able to successfully get a photo as shown in the tutorial, and I can spit out some random coordinates using print(pred_coords), but I can’t find labels pr a description for the coordinates output anywhere to know what coordinate leads to what body point. How do I get meaningful data out of the code?
Similarly, can I adapt the code to read multiple frames or a video file?
I’m currently reading in a video, spitting out multiple frames, and then running the code in a while loop to get a new image with the pose overlay.
Any advice would be helpful. Thanks!