Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7148
Browse
3 results
Search Results
Article Ggnn: Group-Guided Nearest Neighbors for Efficient Image Matching(Springer, 2025) Cine, Ersin; Bastanlar, Yalin; Ozuysal, MustafaThe widely adopted image matching approach remains dependent on exhaustive matching of local features across images. Existing methods aiming to improve efficiency either approximate nearest neighbor (NN) search, compromising accuracy, or apply filtering only after establishing tentative matches, which restricts potential efficiency gains. We challenge the assumption that exhaustive NN search is necessary by proposing a more efficient hierarchical approach that maintains matching accuracy without relying on full-scale NN search. Our key insight is that efficiently identifying sufficiently similar, geometrically meaningful feature matches-rather than the most similar but geometrically random ones-can improve or maintain performance at a lower computational cost. We propose a novel method, Group-Guided Nearest Neighbors (GGNN), which matches groups of features first and then matches individual features only within these matched groups. This hierarchical pipeline reduces the computational complexity of feature matching from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta (n<^>2)$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta (n \sqrt{n})$$\end{document}, significantly improving efficiency. Experimental results on homography estimation demonstrate that GGNN outperforms standard NN search while achieving performance comparable to state-of-the-art methods. Additionally, we formulate GGNN as a general framework, where conventional NN search is a special case with a single global feature group. This formulation provides a continuum of feature matching methods with varying computational costs, enabling automatic selection based on a given time budget.Conference Object Citation - Scopus: 3Derin Öǧrenme ile Zemin Dokusu Sınıflandırma(IEEE, 2018) Ozuysal, MustafaIn this study, we investigate the use of transfer learning on various deep neural network architectures pretained on the ImageNet data set for ground texture classification purposes. We introduce a new ground texture data set collected from seven different areas. We retrain deep neural network's last layer or when possible the full set of layers on this data set. The results show that it is possible to discriminate the ground textures even when very small images are used.Conference Object Robust Keypoint Matching for Three Dimensional Scenes and Object Recognition(IEEE, 2017) Koksal, Ali; Uzyildirim, Furkan Eren; Ozuysal, MustafaIn this paper, we adapt a recently proposed keypoint matching approach for binary descriptors and planar objects to three dimensional objects. We also evaluate the performance of this approach for a museum object recognition application containing more than one hundred paintings. Moreover, we quantify the effect of selecting only descriptors with high matching ratio on the success rate of the object recognition application.