Effect of Parameter Optimization on Classical and Learning-based Image Matching Methods

Download

index.pdf

Date

2021-01-01

Author

Efe, Ufuk
Ince, Kutalmis Gokalp
Alatan, Abdullah Aydın

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

117
views

60
downloads

Deep learning-based image matching methods are improved significantly during the recent years. Although these methods are reported to outperform the classical techniques, the performance of the classical methods is not examined in detail. In this study, we compare classical and learning-based methods by employing mutual nearest neighbor search with ratio test and optimizing the ratio test threshold to achieve the best performance on two different performance metrics. After a fair comparison, the experimental results on HPatches dataset reveal that the performance gap between classical and learning-based methods is not that significant. Throughout the experiments, we demonstrated that SuperGlue is the state-of-the-art technique for the image matching problem on HPatches dataset. However, if a single parameter, namely ratio test threshold, is carefully optimized, a well-known traditional method SIFT performs quite close to SuperGlue and even outperforms in terms of mean matching accuracy (MMA) under 1 and 2 pixel thresholds. Moreover, a recent approach, DFM, which only uses pre-trained VGG features as descriptors and ratio test, is shown to outperform most of the well-trained learning-based methods. Therefore, we conclude that the parameters of any classical method should be analyzed carefully before comparing against a learning-based technique.

URI

https://hdl.handle.net/11511/99632

DOI

https://doi.org/10.1109/iccvw54120.2021.00283

Conference Name

18th IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2021

Collections

Department of Electrical and Electronics Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Automatic Prior Shape Selection for Image Segmentation Guo, Weihong; Qin, Jing; Tarı, Zehra Sibel (2013-07-01) Segmenting images with occluded and missing intensity information is still a difficult task. Intensity based segmentation approaches often lead to wrong results. High vision prior information such as prior shape has been proven to be effective in solving this problem. Most existing shape prior approaches assume known prior shape and segmentation results rely on the selection of prior shape. In this paper, we study how to do simultaneous automatic prior shape selection and segmentation in a variational scheme.
The effect of Laplacian filter in adaptive unsharp masking for infrared image enhancement İLK, HAKKI GÖKHAN; Jane, Onur; İlk Dağ, Özlem (Elsevier BV, 2011-09-01) Image processing, in particular image enhancement techniques have been the focal point of considerable research activity in the last decade. With the aid of an existing image enhancement technique, adaptive unsharp masking (AUM), we propose a novel kernel to be used in AUM filtering in order to enhance discontinuities which occur on the edges of targets of interest in infrared (IR) images. The proposed method uses an adaptive filter approach where an objective function is minimized by using descent algorith...
Effect of the Geometric Inaccuracy in Multivariate Adaptive Regression Spline-based Inverse ECG Solution Approach Onak, O. N.; Serinağaoğlu Doğrusöz, Yeşim; Weber, G. W. (2017-09-27) Spline-based approaches have been applied to inverse problems in several areas. If proper spline bases are chosen, dimension of the problem can be significantly reduced while increasing estimation accuracy and robustness of the inverse procedure. We proposed Multivariate Adaptive Regression Splines (MARS) based methods for the solution of the inverse electrocardiography (ECG) problem considering the temporal and spatial evolution of the epicardial potentials. Our model defines the spline functions in terms ...
SUPER-RESOLUTION RECONSTRUCTION OF HYPERSPECTRAL IMAGES VIA AN IMPROVED MAP-BASED APPROACH Irmak, Hasan; Akar, Gözde; Yuksel, Seniha Esen; Aytaylan, Hakan (2016-07-15) Super-resolution Reconstruction (SRR) is technique to increase the spatial resolution of images. It is especially useful for hyperspectral images (HSI), which have good spectral resolution but low spatial resolution. In this study, we propose an improvement to our previous work and present a novel MAP-MRF (maximum a posteriori-Markov random Fields) based approach for the SRR of HSI. The key point of our approach is to find the abundance maps of an HSI and perform SRR on the abundance maps using MRF based en...
A java toolbox for wavelet based image denoising Tuncer, Güney; Karslıoğlu, Mahmut Onur; Department of Geodetic and Geographical Information Technologies (2006) Wavelet methods for image denoising have became widespread for the last decade. The effectiveness of this denoising scheme is influenced by many factors. Highlights can be listed as choosing of wavelet used, the threshold determination and transform level selection for thresholding. For threshold calculation one of the classical solutions is Wiener filter as a linear estimator. Another one is VisuShrink using global thresholding for nonlinear area. The purpose of this work is to develop a Java toolbox which...

Citation Formats

U. Efe, K. G. Ince, and A. A. Alatan, “Effect of Parameter Optimization on Classical and Learning-based Image Matching Methods,” Virtual, Online, Kanada, 2021, vol. 2021-October, Accessed: 00, 2022. [Online]. Available: https://hdl.handle.net/11511/99632.