View Matching with Blob Features

Per-Erik Forssén, Anders Moe
CRV05, Victoria, BC, Canada
2nd Canadian Conference on Computer and Robot Vision
Pages 228-235
May, 2005

Abstract

This paper introduces a new region based feature for object recognition and image matching. In contrast to many other region based features, this one makes use of colour in the feature extraction stage. We perform experiments on the repeatability rate of the features across scale and inclination angle changes, and show that avoiding to merge regions connected by only a few pixels improves the repeatability. Finally we introduce two voting schemes that allow us to find correspondences automatically, and compare them with respect to the number of valid correspondences they give, and their inlier ratios.

Full Paper

On-line proceedings available on the IEEE Explore website.

Bibtex entry

@inproceedings{fm05,
        Address = {Victoria, BC, Canada},
        Author = {Per-Erik Forss{\'e}n and Anders Moe},
        Booktitle = {2nd Canadian Conference on Computer and Robot Vision},
        Date-Added = {2006-08-08 13:27:59 +0200},
        Date-Modified = {2006-08-08 13:28:22 +0200},
        Month = {May},
        Pages = {228--235},
        Publisher = {{IEEE} Computer Society},
        Title = {View Matching with Blob Features},
        Year = {2005}}