U-CATCH: Using Color ATtribute of image patCHes in binary descriptors

TL;DR

This paper introduces U-CATCH, a method that incorporates color information into binary descriptors by expanding comparisons into RGB and YCbCr spaces, significantly improving matching performance in challenging scenarios.

Contribution

It presents a general, efficient approach to make binary descriptors color-sensitive, enhancing matching accuracy without substantial computational overhead.

Findings

Over 100% improvement in matching accuracy for hard-to-match cases.

Applicable to any binary descriptor for color sensitivity.

Faster than classical descriptors for RGB sampling.

Abstract

In this study, we propose a simple yet very effective method for extracting color information through binary feature description framework. Our method expands the dimension of binary comparisons into RGB and YCbCr spaces, showing more than 100% matching improve ment compared to non-color binary descriptors for a wide range of hard-to-match cases. The proposed method is general and can be applied to any binary descriptor to make it color sensitive. It is faster than classical binary descriptors for RGB sampling due to the abandonment of grayscale conversion and has almost identical complexity (insignificant compared to smoothing operation) for YCbCr sampling.