Depth-based Descriptor for Matching Keypoints in 3D Scenes

• Autorzy: Matusiak K. , Skulimowski P. , Strumillo P.

Identyfikatory

DOI

10.24425/123522

• Języki publikacji: EN

Abstrakty

EN

Keypoint detection is a basic step in many computer vision algorithms aimed at recognition of objects, automatic navigation and analysis of biomedical images. Successful implementation of higher level image analysis tasks, however, is conditioned by reliable detection of characteristic image local regions termed keypoints. A large number of keypoint detection algorithms has been proposed and verified. In this paper we discuss the most important keypoint detection algorithms. The main part of this work is devoted to description of a keypoint detection algorithm we propose that incorporates depth information computed from stereovision cameras or other depth sensing devices. It is shown that filtering out keypoints that are context dependent, e.g. located at boundaries of objects can improve the matching performance of the keypoints which is the basis for object recognition tasks. This improvement is shown quantitatively by comparing the proposed algorithm to the widely accepted SIFT keypoint detector algorithm. Our study is motivated by a development of a system aimed at aiding the visually impaired in space perception and object identification.

Słowa kluczowe

EN

feature matching; keypoints detection; object recognition; depth map

• Wydawca: Polish Academy of Sciences, Committee of Electronics and Telecommunication
• Czasopismo: International Journal of Electronics and Telecommunications
• Rocznik: 2018
• Tom: Vol. 64, No. 3
• Strony: 299--306
• Opis fizyczny: Bibliogr. 20 poz., il., rys., wykr

Twórcy

autor

Matusiak K.

• Lodz University of Technology, Institute of Electronics, Lodz, Poland

autor

Skulimowski P.

• Lodz University of Technology, Institute of Electronics, Lodz, Poland

autor

Strumillo P.

• Lodz University of Technology, Institute of Electronics, Lodz, Poland

Bibliografia

• [1] T. Tuytelaars and K. Mikolajczyk, Local Invariant Feature Detectors: A Survey, Foundations and Trends in Computer Graphics and Vision Volume 3, Issue 3, pp. 177-280 (2007).
• [2] ZED Stereo Camera, www.stereolabs.com, accessed 2018.03.27.
• [3] Google Tango Project, get.google.com/tango, accessed 2018.03.27.
• [4] Structure Sensor, structure.io, accessed 2018.03.27.
• [5] D. R. dos Santos, M. A. Basso, K. Khoshelham, E. de Oliveira, N. L. Pavan, G. Vosselman, Mapping Indoor Spaces by Adaptive Coarseto-Fine Registration of RGB-D Data, IEEE Geoscience and Remote Sensing Letters, Volume 13, Issue 2 (2016).
• [6] O. Wasenmller, M. Meyer, D. Stricker, CoRBS: Comprehensive RGBD benchmark for SLAM using Kinect v2, IEEE Winter Conference on Applications of Computer Vision (2016).
• [7] M. Karpushin, G. Valenzise and F. Dufaux, Improving distinctivness of BRISK features using depth maps, IEEE International Conference on Image Processing (2015).
• [8] M. Bujacz, P. Skulimowski, and P. Strumillo. Naviton - a prototype mobility aid for auditory presentation of three-dimensional scenes to the visually impaired. J. Audio Eng. Soc, 60(9):696–708, 2012.
• [9] K. Matusiak, P. Skulimowski, and P. Strumillo. Object recognition in a mobile phone application for visually impaired users. In 2013 6th International Conference on Human System Interactions (HSI), pages 479–484, June 2013.
• [10] K. Matusiak, P. Skulimowski, P. Strumillo, Unbiased evaluation of keypoint detectors with respect to rotation invariance, IET Computer Vision, Volume 11, Issue 7, 507-516 (2017).
• [11] D. G. Lowe, Distinctive Image Features from Scale-Invariant Keypoints, International Journal of Computer Vision, Volume 60, Issue 2, 91-110 (2004).
• [12] H. Bay, A. Ess, T. Tuytelaars, L. Van Gool, Speeded-Up Robust Features (SURF), Computer Vision and Image Understanding, Volume 110, Issue 3, 346-359 (2008).
• [13] E. Rublee, V. Rabaud, K. Konolige, G. Bradski, ORB: an efficient alternative to SIFT or SURF, IEEE International Conference on Computer Vision 2011.
• [14] B. Steder, R. Bogdan, R. Kurt and K. W. Burgard, NARF: 3D Range Image Features for Object Recognition, Workshop on Defining and Solving Realistic Perception Problems in Personal Robotics at the IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (2010).
• [15] S. Leutenegger, M. Chliand and R. Y.Siegwart, BRISK: Binary Robust invariant scalable keypoints, IEEE International Conference on Computer Vision, pp. 2548-2555 (2011).
• [16] E. Rosten and T. Drummond, Machine Learning for High-Speed Corner Detection, Computer Vision ECCV 2006, Volume 1, pp. 430-443 (2006).
• [17] S. Martull, M. P. Martorell, K. Fukui, Realistic CG Stereo Image Dataset with Ground Truth Disparity Maps, ICPR2012 workshop Trak-Mark2012, pp.40-42 (2012).
• [18] C. Choi, A. J. B. Trevor, H. I. Christensen, RGB-D edge detection and edge-based registration, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems (2013).
• [19] Q. Yu, J. Liang, J. Xiao, H. Lu, Z. Zheng, A Novel perspective invariant feature transform for RGB-D images, Computer Vision and Image Understanding, 167, 109-120 (2018).
• [20] E. R. Nascimento, G. L. Oliveira, M. F. M. Campos, A. W. Vieira On the development of a robust, fast, lightweight keypoint descriptor, Neurocomputing, 120, 141-155 (2013).

Uwagi

1. Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2018).

2. This work was partially supported by the European Unions Horizon 2020 Research and Innovation Programme under grant agreement No 643636 Sound of Vision.