Can machine vision map humans from videos to 3D Models? Yes! DensePose is a new architecture by the team at Facebook AI research that does just that. It uses a convolutional network with some special features like region of interest pooling and cascading to make this happen. It was also trained on a newly created labeled dataset that mapped human poses to 3D models. The team open sourced the dataset but not the code, but using the details in the paper we can recreate their results. I'll explain how it works in this video.
Code for this video:
https://github.com/llSourcell/3D_Pose_Estimation
Please Subscribe! And like. And comment. That's what keeps me going.
Want more education? Connect with me here:
Twitter: https://twitter.com/sirajraval
Facebook: https://www.facebook.com/sirajology
instagram: https://www.instagram.com/sirajraval
More learning resources:
https://arxiv.org/abs/1802.00434
http://densepose.org/
https://www.youtube.com/watch?v=pW6nZXeWlGM&t=157s
https://www.youtube.com/watch?v=2CWVgTNremI
https://www.youtube.com/watch?v=NnzzSkKKoa8
https://www.youtube.com/watch?v=hyHNpl0xSkw
Join us in the Wizards Slack channel:
http://wizards.herokuapp.com/
And please support me on Patreon:
https://www.patreon.com/user?u=3191693
