Learning to Compare Image Patches via Convolutional Neural Networks

TL;DR

This paper introduces a CNN-based approach to learn a general similarity function for comparing image patches directly from raw image data, improving performance on various computer vision benchmarks.

Contribution

It presents a novel CNN architecture for image patch comparison that outperforms existing methods and learns directly from image data without manual feature design.

Findings

Significant performance improvements over state-of-the-art methods

Effective CNN architectures tailored for patch comparison

Robustness to various image appearance changes

Abstract

In this paper we show how to learn directly from image data (i.e., without resorting to manually-designed features) a general similarity function for comparing image patches, which is a task of fundamental importance for many computer vision problems. To encode such a function, we opt for a CNN-based model that is trained to account for a wide variety of changes in image appearance. To that end, we explore and study multiple neural network architectures, which are specifically adapted to this task. We show that such an approach can significantly outperform the state-of-the-art on several problems and benchmark datasets.