Part Affinity Fields | Human Pose Estimation | Computer Vision on Humans

preview_player
Показать описание
Estimating Human and Hand Poses using a technique called Part Affinity Fields. The Part Affinity Fields use stages of a Convolutional Neural Network to predict key points and the vector fields joining those key points. The key points from different humans are grouped into separate skeletons by approximating the solution to a Maximum Bipartite Matching Problem.

Feel free to leave a comment or message me on Twitter/LinkedIn in case of any questions, doubts, suggestions or improvements.

Links
Рекомендации по теме
Комментарии
Автор

Thanks for the video! Cleared lot of my doubts 😀

kimi
Автор

Unable to download the weights. Can you pls help.

sohomsen
Автор

I don’t understand where is the part related to the PAFs in the code? Where is the implementation of it?

naiyraelkady
Автор

Unable to download the weights. Can you pls help.

luuminhquan