Improvement of ASIFT for Object Matching Based on Optimized Random Sampling
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This paper proposes an efficient matching algorithm based on ASIFT (Affine Scale-Invariant Feature Transform) which is fully invariant to affine transformation. In our approach, we proposed a method of reducing similar measure matching cost and the number of outliers. First, we combined the Manhattan and Chessboard metrics replacing the Euclidean metric by a linear combination for measuring the similarity of keypoints. These two metrics are simple but really efficient. Using our method the computation time for matching step was saved and also the number of correct matches was increased. By applying an Optimized Random Sampling Algorithm (ORSA), we can remove most of the outlier matches to make the result meaningful. This method was experimented on various combinations of affine transform. The experimental result shows that our method is superior to SIFT and ASIFT.
Abstract
This paper proposes an efficient matching algorithm based on ASIFT (Affine Scale-Invariant Feature Transform) which is fully invariant to affine transformation. In our approach, we proposed a method of reducing similar measure matching cost and the number of outliers. First, we combined the Manhattan and Chessboard metrics replacing the Euclidean metric by a linear combination for measuring the similarity of keypoints. These two metrics are simple but really efficient. Using our method the computation time for matching step was saved and also the number of correct matches was increased. By applying an Optimized Random Sampling Algorithm (ORSA), we can remove most of the outlier matches to make the result meaningful. This method was experimented on various combinations of affine transform. The experimental result shows that our method is superior to SIFT and ASIFT.
참고문헌 (16)
- J.-M. Morel, "ASIFT: A New Framework for Fully Affine Invariant Image Comparison", SIAM Journal on Imaging Sciences, 2, pp.438~469 (2009)
- D.G. Lowe, "Distinctive image features from scale-invariant keypoints", International Journal of Computer Vision, 60(2), pp.91~110 (2004)
- J.M. Morel, "Is SIFT scale invariant?", Inverse Problems and Imaging, 5(1), pp.115~136 (2011)
- On the Consistency of the SIFT Method, "J. Morel", CMLA, ENSCachan, ~ (2008)
- "ETHZ Toys V 1.0 Dataset"[URL]
- A. Desolneux, "From Gestalt Theory to Image Analysis: A Probabilistic Approach" (2008), Springer
- J. Zhang, "Local features and kernels for classification of texture and object categories: A comprehensive study", Int. J. Comput. Vision, 73(2), pp.213~238 (2007)
- V. Ferrari, "Simultaneous object recognition and segmentation from single or multiple model views", Int. J. Comput.Vision, 67(2), pp.159~188 (2006)
- N. Snavely, "Modeling the world from internet photo collections", International Journal of Computer Vision (2008)
- A. Bosch, "Scene classification using a hybrid generative", discriminative approach, 2008, pp.30(4), 712~727
- M. Brown, "Automatic panoramic image stitching using invariant features", International Journal of Computer Vision, 74(1), pp.59~73 (2007)
- J. Jia, "Image stitching using structure deformation", IEEE Transactions on Pattern Analysis and Machine Intelligence, 30(4), pp.617~631 (2008)
- J. Matas, "Robust wide baseline stereo from maximally stable extremal regions", British Machine Vision Conference, pp.384~393 (2002)
- R. Deriche, "Robust recovery of the epipolar geometry for an uncalibrated stereo rig", Proc. European Conference on Computer Vision, pp.567~576 (1994)
- A. Kushal, "Modeling 3D objects from stereo view and recognizing them in photographs", Proc. European Conference on Computer Vision (2006)
- "Morel Yu’s Dataset"[URL]