Novel shrinking residual convolutional neural network for efficient accurate stereo matching |
| |
| Affiliation: | 1. School of Electronic Information, Wuhan University, Wuhan, China;2. School of Computers, Guangdong University of Technology, Guangzhou, China;3. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, China;1. Department of Mathematics and Physics, North China Electric Power University, Baoding, Hebei 071003, China;2. Science and Technology on Reactor System Design Technology Laboratory, Nuclear Power Institute of China, Chengdu 610213, China;1. University of São Paulo, Institute of Mathematics and Statistics, São Paulo, SP, Brazil;2. Universidade Federal de São Paulo, Instituto de Ciência e Tecnologia, São José dos Campos, SP, Brazil;3. University of Campinas, Institute of Computing, Campinas, SP, Brazil;1. Department of Electronic and Information Engineering, Beijing Jiaotong University, Beijing 100044, China;2. College of information, Liaoning University, Liaoning 110036, China |
| |
| Abstract: | For stereo matching based on patch comparing using convolutional neural networks (CNNs), the matching cost estimation is highly dependent on the network structure, and the patch comparing is time consuming for traditional CNNs. Accordingly, we propose a stereo matching method based on a novel shrinking residual CNN, which consists of convolutional layers and skip-connection layers, and the size of the fully connected layers decreases progressively. Firstly, a layer-by-layer shrinking size model is adopted for the full-connection layers to greatly increase the running speed. Secondly, the convolutional layer and the residual structure are fused to improve patch comparing. Finally, the Loss function is re-designed to give higher weights to hard-classified examples compared with the standard cross entropy loss. Experimental results on KITTI2012 and KITTI2015 demonstrate that the proposed method can improve the operation speed while maintaining high accuracy. |
| |
| Keywords: | Stereo matching Matching cost Residual convolutional neural network |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
