On the Comparison of Classic and Deep Keypoint Detector and Descriptor Methods

TL;DR

This paper compares the performance of classic hand-crafted and recent deep keypoint detector and descriptor methods across various tasks and datasets, highlighting their relative strengths and execution times.

Contribution

It provides a comprehensive benchmark of classic and deep keypoint methods, including new deep models LF-Net and SuperPoint, on diverse real-world images.

Findings

Some classic methods outperform deep models in accuracy.

SuperPoint is the fastest keypoint detector-descriptor.

Certain classic and deep approaches are comparably effective.

Abstract

The purpose of this study is to give a performance comparison between several classic hand-crafted and deep key-point detector and descriptor methods. In particular, we consider the following classical algorithms: SIFT, SURF, ORB, FAST, BRISK, MSER, HARRIS, KAZE, AKAZE, AGAST, GFTT, FREAK, BRIEF and RootSIFT, where a subset of all combinations is paired into detector-descriptor pipelines. Additionally, we analyze the performance of two recent and perspective deep detector-descriptor models, LF-Net and SuperPoint. Our benchmark relies on the HPSequences dataset that provides real and diverse images under various geometric and illumination changes. We analyze the performance on three evaluation tasks: keypoint verification, image matching and keypoint retrieval. The results show that certain classic and deep approaches are still comparable, with some classic detector-descriptor combinations overperforming pretrained deep models. In terms of the execution times of tested implementations, SuperPoint model is the fastest, followed by ORB. The source code is published on \url{https://github.com/kristijanbartol/keypoint-algorithms-benchmark}.