Improved learning of I2C distance and accelerating the neighborhood search for image classification |
| |
| Authors: | Zhengxiang Wang Yiqun Hu Liang-Tien Chia |
| |
| Affiliation: | 1. Center for Multimedia and Network Technology, School of Computer Engineering, Nanyang Technological University, 639798 Singapore, Singapore;2. School of Computer Science and Software Engineering, The University of Western Australia, Australia;1. Department of Civil Engineering, Shanghai University, Shanghai 200444, China;2. College of Textiles, Donghua University, Shanghai 201620, China;3. Engineering Research Center of Technical Textiles, Ministry of Education, Donghua University, Shanghai 201620, China;4. School of Urban Rail Transportation, Shanghai University of Engineering Science, Shanghai, 201620, China;1. Launch Complex Team (LCT), KSLV-II R&D Program Executive Office, Korea Aerospace Research Institute (KARI), 169-84 Gwahak-ro, Yuseong-gu, Daejeon 305-806, Republic of Korea;2. The William E. Boeing Department of Aeronautics and Astronautics, University of Washington, USA;3. Body Durability CAE Team, Automotive Research & Development Division, Hyundai Motor Group, 150 Hyundai Yeonguso-ro, Hwaseong-si, Gyeonggi-do 445-706, Republic of Korea;4. School of Mechanical, Aerospace and Systems Engineering, KAIST, 291 Daehak-ro, Yuseong-gu, Daejeon 305-701, Republic of Korea;1. Mechanics of Intelligent Structures Department, Institute of Fluid-Flow Machinery, Polish Academy of Sciences, 14 Fiszera Street, 80-231 Gdansk, Poland;2. Faculty of Automotive and Construction Machinery, Warsaw University of Technology, 84 Narbutta Street, 02-524 Warsaw, Poland |
| |
| Abstract: | Image-to-class (I2C) distance is a novel measure for image classification and has successfully handled datasets with large intra-class variances. However, due to the lack of a training phase, the performance of this distance is easily affected by irrelevant local features that may hurt the classification accuracy. Besides, the success of this I2C distance relies heavily on the large number of local features in the training set, which requires expensive computation cost for classifying test images. On the other hand, if there are small number of local features in the training set, it may result in poor performance.In this paper, we propose a distance learning method to improve the classification accuracy of this I2C distance as well as two strategies for accelerating its NN search. We first propose a large margin optimization framework to learn the I2C distance function, which is modeled as a weighted combination of the distance from every local feature in an image to its nearest-neighbor (NN) in a candidate class. We learn these weights associated with local features in the training set by constraining the optimization such that the I2C distance from image to its belonging class should be less than that to any other class. We evaluate the proposed method on several publicly available image datasets and show that the performance of I2C distance for classification can significantly be improved by learning a weighted I2C distance function. To improve the computation cost, we also propose two methods based on spatial division and hubness score to accelerate the NN search, which is able to largely reduce the on-line testing time while still preserving or even achieving a better classification accuracy. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
