A Comparative Evaluation of Well-known Feature Detectors and Descriptors

Year 2015, Volume: 3 Issue: 1, 1 - 6, 17.01.2015

Şahin Işık

https://doi.org/10.18100/ijamec.60004

Cited By: 38

Abstract

Comparison of feature detectors and descriptors and assessing their performance is very important in computer vision. In this study, we evaluate the performance of seven combination of well-known detectors and descriptors which are SIFT with SIFT, SURF with SURF, MSER with SIFT, BRISK with FREAK, BRISK with BRISK, ORB with ORB and FAST with BRIEF. The popular Oxford dataset is used in test stage. To compare the performance of each combination objectively, the effects of JPEG compression, zoom and rotation, blur, viewpoint and illumination variation have investigated in terms of precision and recall values. Upon inspecting the obtained results, it is observed that the combination of ORB with ORB and MSER with SIFT can be preferable almost in all possible situations when the precision and recall results are considered. Moreover, the speed of FAST with BRIEF is superior to others.

Keywords

Image matching; feature detector; feature descriptors; SIFT; SURF; BRIEF; FAST; BRISK; ORB; MSER

Original Research Paper

Abstract

References

• Peng Z. Efficient matching of robust features for embedded SLAM, 2012.
• Heinly J. Dunn E., and Frahm J.-M. Comparative evaluation of binary features, Computer Vision–ECCV, Springer, 2012, pp. 759-773.
• El-gayar M. and Soliman H. A comparative study of image low level feature extraction algorithms, Egyptian Informatics Journal, Vol.14, Number 2, 2013, pp. 175-181.
• Lingua A. Marenchino D. and Nex F. Performance analysis of the SIFT operator for automatic feature extraction and matching in photogrammetric applications, Sensors, Vol. 9, Number 5, 2009, pp. 3745-3766.
• Schaeffer C. A comparison of keypoint descriptors in the context of pedestrian detection: freak vs. surf vs. brisk.
• Oxford Dataset, robots.ox.ac.uk/~vgg/data/data-aff.html
• Lowe, D. G. Object recognition from local scale-invariant features. Computer vision, Vol. 2, 1999, pp. 1150-1157.
• Lowe D.G. Distinctive image features from scale-invariant keypoints, International journal of computer vision, Vol. 60, Number 2, 2004, pp. 91-110.
• Bay H. Ess A. Tuytelaars T. and Van Gool L. Speeded-up robust features (SURF), Computer vision and image understanding, Vol. 110, Number 3, 2008, pp. 346-359.
• Rosten E. and Drummond T. Machine learning for high-speed corner detection, Computer Vision (ECCV 2006), Springer, 2006, pp. 430-443
• Rosten E. Porter R. and Drummond, T. Faster and better: A machine learning approach to corner detection, Pattern Analysis and Machine Intelligence, IEEE Transactions on, Vol. 32, Number 1, 2010, pp. 105-119.
• http://www.edwardrosten.com/work/fast.html
• Leutenegger S. Chli M. and Siegwart R. Y. BRISK: Binary robust invariant scalable keypoints. Computer Vision (ICCV), IEEE, 2011, pp. 2548-2555.
• Matas, J., Chum, O., Urban, M., and Pajdla, T. Robust wide-baseline stereo from maximally stable extremal regions, Image and vision computing, Vol. 22, Number 10, 2004, pp. 761-767.
• Nistér D. and Stewénius H. Linear time maximally stable extremal regions: ‘Computer Vision (ECCV 2008), Springer, 2008, pp. 183-196.
• Obdržálek D. Basovník S. Mach L. and Mikulík A. Detecting scene elements using maximally stable colour regions, Research and Education (Robotics-EUROBOT 2009), Springer, 2010, pp. 107-115.
• Rublee E. Rabaud V. Konolige K. and Bradski G. ORB: an efficient alternative to SIFT or SURF. (Computer Vision (ICCV)), IEEE, 2011, pp. 2564-2571.
• Alahi A. Ortiz R. and Vandergheynst P. Freak: Fast retina keypoint. Computer Vision and Pattern Recognition (CVPR), IEEE, 2012, p. 510-517.
• Mikolajczyk K. and Schmid C. A performance evaluation of local descriptors, Pattern Analysis and Machine Intelligence, 2005, Vol. 27, Number 10, pp. 1615-1630.