Results are presented for all 5 benchmark datasets. For each dataset, they are divided between “fixed-scale” and “adaptive-scale” detectors (see [1,2] for more details).
Algorithms currently present in our evaluation are:
Fixed-scale:
- Intrinsic Shape Signatures (ISS, red) [3]
- Local Surface Patches (LSP, green) [4]
- KeyPoint Quality (KPQ, blue) [5]
- Heat Kernel Signatures (HKS, brown) [6]
Adaptive-scale:
- Laplace-Beltrami Scale Space (LBSS, pink crossed) [7]
- MeshDoG (yellow crossed) [8]
- KeyPoint Quality Adaptive Scale (KPQ-AS, light blue) [5]
- Salient Points (SP, black striped) [9]
References
[3] Y. Zhong, “Intrinsic shape signatures: A shape descriptor for 3D object recognition”, Proc. Int. Conf. on Computer Vision Workshops, 2009
[4] H. Chen H, B. Bhanu, “3D free-form object recognition in range images using local surface patches”, Pattern Recognition Letters 28(10):1252-1262, 2007
[5] A. Mian, M. Bennamoun, R. Owens, “On the repeatability and quality of keypoints for local feature-based 3D object retrieval from cluttered scenes”, Int. Journal of Computer Vision, 89(2-3):348-361, 2010
[6] J. Sun, M. Ovsjanikov, L. Guibas, “A concise and provably informative multi-scale signature based on heat diffusion”, Proc. Eurographics Symposium on Geometry Processing (SGP), pp 1383-1392, 2009
[7] R. Unnikrishnan, M. Hebert, “Multi-scale interest regions from unorganized point clouds”, Proc. Workshop on Search in 3D (S3D), pp 1-8, 2008
[8] A. Zaharescu, E. Boyer, K. Varanasi, R. Horaud, “Surface feature detection and description with applications to mesh matching”, Proc Int Conf on Computer Vision and Pattern Recognition (CVPR), pp. 373-380, 2009
[9] U. Castellani, M. Cristani, S. Fantoni, “Sparse points matching by combining 3D mesh saliency with statistical descriptors”, Proc Computer Graphics Forum, pp 643-652, 2008