HOG is a feature descriptor used in computer vision and image processing for object detection. It counts occurrences of gradient orientation in localized portions of an image, and was first described by Robert K. McConnell in 1986. It became widely used in 2005 when Navneet Dalal and Bill Triggs presented their work on HOG descriptors at the CVPR conference.
University of Washington
Winter 2022
A general introduction to computer vision, this course covers traditional image processing techniques and newer, machine-learning based approaches. It discusses topics like filtering, edge detection, stereo, flow, and neural network architectures.
No concepts data
+ 24 more concepts