From handcrafted to deep local features

TL;DR

This paper reviews the evolution of local features from handcrafted techniques to deep learning methods, emphasizing their role in 3D reconstruction and discussing benchmarks, challenges, and future directions.

Contribution

It provides a comprehensive chronological overview of local feature extraction methods, highlighting key challenges and the transition to deep learning approaches in computer vision.

Findings

Deep learning has significantly advanced local feature extraction.

Handcrafted features are still relevant for understanding modern methods.

Benchmark evaluations reveal strengths and limitations of different approaches.

Abstract

This paper presents an overview of the evolution of local features from handcrafted to deep-learning-based methods, followed by a discussion of several benchmarks and papers evaluating such local features. Our investigations are motivated by 3D reconstruction problems, where the precise location of the features is important. As we describe these methods, we highlight and explain the challenges of feature extraction and potential ways to overcome them. We first present handcrafted methods, followed by methods based on classical machine learning and finally we discuss methods based on deep-learning. This largely chronologically-ordered presentation will help the reader to fully understand the topic of image and region description in order to make best use of it in modern computer vision applications. In particular, understanding handcrafted methods and their motivation can help to understand modern approaches and how machine learning is used to improve the results. We also provide references to most of the relevant literature and code.