论文标题
关于经典和深关键点检测器和描述符方法的比较
On the Comparison of Classic and Deep Keypoint Detector and Descriptor Methods
论文作者
论文摘要
这项研究的目的是在几种经典的手工制作和深度关键点检测器和描述符方法之间进行性能比较。特别是，我们考虑以下经典算法：SIFT，SUFT，ORB，快速，快速，MSER，HARRIS，HARRIS，HARRIS，AKAZE，AKAZE，AGAST，GFTT，GFTT，FREAK，FREAK，SHICOTSIFT，其中所有组合的子集都与检测器描述器中的管道配对。此外，我们分析了两个最近和透视深度检测器描述符模型LF-NET和SUPERPOINT的性能。我们的基准测试依赖于HP序列数据集，该数据集在各种几何和照明变化下提供了真实和多样的图像。我们分析了三个评估任务的性能：关键点验证，图像匹配和关键点检索。结果表明，某些经典和深度方法仍然是可比性的，一些经典的检测器描述符组合表现出了预估计的深层模型。就测试实现的执行时间而言，SuperPoint模型是最快的，其次是Orb。源代码在\ url {https://github.com/kristijanbartol/keypoint-algorithms-benchmark}上发布。
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}.
