Learning context-sensitive similarity by shortest path propagation |
| |
| Authors: | Jingyan Wang Yongping Li Xiang Bai Ying Zhang Chao Wang Ning Tang |
| |
| Affiliation: | 1. Shanghai Institute of Applied Physics, Chinese Academy of Sciences, 2019 Jialuo Road, Shanghai 201800, PR China;2. Department of Electronics and Information Engineering, Huazhong University of Science and Technology, 1037 Luoyu Road, Wuhan, Hubei Province 430074, PR China;3. OGI School of Science and Engineering, Oregon Health & Science University (OHSU), Beaverton, OR 97006, US;1. EPFL, Switzerland;2. Center for Research & Technology Hellas (CERTH), Greece;1. Dept. Electrical and Computer Engineering, Tufts University, Medford, MA, USA;2. Dept. Radiology, Brigham and Women’s Hospital, Boston, MA, USA;3. Dept. Neurobiology, Harvard Medical School, Boston, MA, USA;1. Department of Computer Science and Medical Equipment, Faculty of Biomedical Engineering, Silesian University of Technology, ul. Roosevelta 40, 41-800 Zabrze, Poland;2. Research Group for Pattern Recognition, University of Siegen, Hoelderlinstrasse 3, D-57076 Siegen, Germany;3. University of Applied Sciences Koblenz, Department of Mathematics and Technology, Joseph-Rovan-Allee 2, 53424 Remagen, Germany;4. Herz-Jesu Hospital Department of Diagnostic and Interventional Radiology and Nuclear Medicine, Südring 8, 56428 Dernbach, Germany;1. Engineering Training Center, Shenyang Aerospace University, Shenyang 110136, China;2. School of Computer, Shenyang Aerospace University, Shenyang 110136, China |
| |
| Abstract: | In this paper, we introduce a novel shape/object retrieval algorithm shortest path propagation (SSP). Given a query object q and a target database object p, we explicitly find the shortest path between them in the distance manifold of the database objects. Then a new distance measure between q and p is learned based on the database objects on the shortest path to replace the original distance measure. The promising results on both MEPG-7 shape dataset and a protein dataset demonstrate that our method can significantly improve the ranking of the object retrieval. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
