sendeniz/sift-ransac-bow

Salient Feature Extraction for Image Classification

General:
This repo contains a C++ implementation of Scale Invariant Feature Transform (SIFT) and Random sample consensus (RANSAC) for image similarity and image classification using a Bag of Words (BoW) classifier. Salient feature are visualised and matching features are shown before and after RANSAC.

Fig.1 - SIFT salient features visualised on the Cathedral of the Holy Cross in Boston.

Fig.2 - SIFT matches between the Notre Dame Cathedral and the Holy Cross on the left and after RANSAC on the right .