Constructing Binary Descriptors with a Stochastic Hill Climbing Search

TL;DR

This paper introduces a stochastic hill climbing method for constructing binary image patch descriptors, demonstrating superior discriminative power over recent alternatives with a simple, fast, and parameter-free approach.

Contribution

The paper presents a novel stochastic hill climbing technique for binary descriptor construction that outperforms recent methods on standard benchmarks.

Findings

Outperforms recent binary descriptor methods on discriminative benchmarks.

Simple, fast, and parameter-free method.

Effective for image patch representation.

Abstract

Binary descriptors of image patches provide processing speed advantages and require less storage than methods that encode the patch appearance with a vector of real numbers. We provide evidence that, despite its simplicity, a stochastic hill climbing bit selection procedure for descriptor construction defeats recently proposed alternatives on a standard discriminative power benchmark. The method is easy to implement and understand, has no free parameters that need fine tuning, and runs fast.