Light weight stereo matching via deep extraction and integration of low and high level information

Authors

Xianzhe Xu, Tianjin University
Yonghong Hou, Tianjin UniversityFollow
Pichao Wang, Alibaba Group (U. S.) Inc., University of WollongongFollow
Zhongyu Jiang, Tianjin University
Wanqing Li, University of WollongongFollow

RIS ID

138458

Publication Details

Xu, X., Hou, Y., Wang, P., Jiang, Z. & Li, W. (2019). Light weight stereo matching via deep extraction and integration of low and high level information. IEEE International Conference on Multimedia and Expo (ICME) 2019 (pp. 320-325). United States: IEEE.

Abstract

Deep convolutional neural networks (CNN) have demonstrated remarkable progress in stereo matching recently. However, disparity estimation in the ill-posed regions is still difficult. In addition, CNN based stereo matching methods often have impractical computational complexity and memory consumption. To address these problems we propose an end-to-end light weight CNN architecture to effectively learn and integrate low and high level information. To achieve this, a novel enhancement block built upon group convolution and dilated-convolution is proposed. Compared with state-of-the-art methods, the proposed method achieved competitive performance with the least number of network parameters on the Flyingthings3d and KITTI datasets.