Local-Feature-Matching-using-Harris-Corner-Detector-and-SIFT-Descriptor-

This repo contains a fully developed code for Local Feature Matching using Harris Corner Detector and SIFT Descriptor

This project was developed for Computer Vision class Spring 2019.

Table of content

• Dataset
• Harris Corner 3. Results
  1. Algorithm
• SIFT Descriptor
  • Results
  • Algorithm
• Feature Matching
  • Results
  • Algorithm

Dataset

The dataset is simply 6 images that can be considered as 3 pairs of images as following:

1. Notre Dame de Paris
2. Mount Rushmore
3. Gaudi's Episcopal Palace

Harris Corner

Results

Hybrid images can be constructed by using 2 images with respectable shapes and using a low pass filter on one image and a high pass image on the other one.

Algorithm

You can find the algorithm expalined thoroughly in the report.

SIFT Descriptor

Results

The SIFT descriptor output a matrix k*n where k is number of interest points and n is the feature vector for each interest point.

Algorithm

You can find the algorithm explained thoroughly in the report.

Feature Matching

Results

Reached Accruracy of 99%

Algorithm

You can find the algorithm explained thoroughly in the report.