In Lecture 3 Jeremy showed some pictures from someone(s) working on a project to predict the coordinates of a face in a picture. Can someone point me to some information about this?
If I understand your question correctly, that is a problem called “object detection”. Jeremy talked about the MS COCO dataset which is one of the standard object detection datasets. Google just came out with a large new one I believe.
Some of the state of the art algorithms for object detection include YOLOv2. You can look at past sessions of fastai Part 2 and Stanford cs231n for good introductions.
I think you’re looking for this https://data.vision.ee.ethz.ch/cvl/gfanelli/head_pose/head_forest.html#db
1 Like
This is it. thank you.