About this document
Hough-transform for Object Recognition by over related2 is a document available to read on EtoBox.
This document proposes an object recognition approach that uses SURF features extracted from RGB images and clusters them in Hough space to identify object poses. It describes extracting SURF features from training images of objects, matching them to features from test images using nearest neighbor matching, clustering the matched features in Hough space to generate pose hypotheses, and verifying poses using homography. The approach is applied in the RoboCup@Home league and the software is provided as an op
- Author
- over related2
- Language
- EN