Image Matching by Bare Homography

被引:3
作者
Bellavia, Fabio [1 ]
机构
[1] Univ Palermo, Dept Math & Comp Sci, Palermo, Italy
关键词
Image matching; Pipelines; Artificial neural networks; Feature extraction; Training; Transmission line matrix methods; Simultaneous localization and mapping; Keypoint matching; local planar homography; affine constraints; SIFT; LoFTR; SuperGlue; AdaLAM; RANSAC; FEATURES; SCALE;
D O I
10.1109/TIP.2023.3346682
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper presents Slime, a novel non-deep image matching framework which models the scene as rough local overlapping planes. This intermediate representation sits in-between the local affine approximation of the keypoint patches and the global matching based on both spatial and similarity constraints, providing a progressive pruning of the correspondences, as planes are easier to handle with respect to general scenes. Slime decomposes the images into overlapping regions at different scales and computes loose planar homographies. Planes are mutually extended by compatible matches and the images are split into fixed tiles, with only the best homographies retained for each pair of tiles. Stable matches are identified according to the consensus of the admissible stereo configurations provided by pairwise homographies. Within tiles, the rough planes are then merged according to their overlap in terms of matches and further consistent correspondences are extracted. The whole process only involves homography constraints. As a result, both the coverage and the stability of correct matches over the scene are amplified, together with the ability to spot matches in challenging scenes, allowing traditional hybrid matching pipelines to make up lost ground against recent end-to-end deep matching methods. In addition, the paper gives a thorough comparative analysis of recent state-of-the-art in image matching represented by end-to-end deep networks and hybrid pipelines. The evaluation considers both planar and non-planar scenes, taking into account critical and challenging scenarios including abrupt temporal image changes and strong variations in relative image rotations. According to this analysis, although the impressive progress done in this field, there is still a wide room for improvements to be investigated in future research.
引用
收藏
页码:696 / 708
页数:13
相关论文
共 65 条
  • [1] A Comparative Study of Interest Point Performance on a Unique Data Set
    Aanaes, Henrik
    Dahl, Anders Lindbjerg
    Pedersen, Kim Steenstrup
    [J]. INTERNATIONAL JOURNAL OF COMPUTER VISION, 2012, 97 (01) : 18 - 35
  • [2] Arandjelovic R, 2012, PROC CVPR IEEE, P2911, DOI 10.1109/CVPR.2012.6248018
  • [3] HPatches: A benchmark and evaluation of handcrafted and learned local descriptors
    Balntas, Vassileios
    Lenc, Karel
    Vedaldi, Andrea
    Mikolajczyk, Krystian
    [J]. 30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, : 3852 - 3861
  • [4] A Large-Scale Homography Benchmark
    Barath, Daniel
    Mishkin, Dmytro
    Polic, Michal
    Forstner, Wolfgang
    Matas, Jiri
    [J]. 2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2023, : 21360 - 21370
  • [5] Finding Geometric Models by Clustering in the Consensus Space
    Barath, Daniel
    Rozumnyi, Denys
    Eichhardt, Ivan
    Hajder, Levente
    Matas, Jiri
    [J]. 2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR, 2023, : 5414 - 5424
  • [6] Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filters
    Barroso-Laguna, Axel
    Riba, Edgar
    Ponsa, Daniel
    Mikolajczyk, Krystian
    [J]. 2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, : 5835 - 5843
  • [7] Challenges in Image Matching for Cultural Heritage: An Overview and Perspective
    Bellavia, F.
    Colombo, C.
    Morelli, L.
    Remondino, F.
    [J]. IMAGE ANALYSIS AND PROCESSING, ICIAP 2022 WORKSHOPS, PT I, 2022, 13373 : 210 - 222
  • [8] Bellavia F., 2022, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, V46, P73
  • [9] Bellavia F., 2023, PROCIMAGE ANAL WORKS
  • [10] HarrisZ+: Harris corner selection for next-gen image matching pipelines
    Bellavia, Fabio
    Mishkin, Dmytro
    [J]. PATTERN RECOGNITION LETTERS, 2022, 158 : 141 - 147