A fast HOG feature extractor is implemented in this repo. This code for extracting HOG features in images is a naive implementation based on the work of The Fastest Deformable Part Model for Object Detection, which is published in CVPR2014. Note that only the feature extraction part is implemented.
It was every interesting that I wrote this HOG extractor on a Saturday in 2014. My roomate went to his company and locked me home. I could not open the door, having nothing to do. Then I thought why not implementing this fast HOG extractor. So here it is.
| Input | Output |
|---|---|
 |
 |
Yan, Junjie, et al. "The Fastest Deformable Part Model for Object Detection." IEEE Conference on Computer Vision and Pattern Recognition IEEE, 2014:2497-2504.