季顺平, 罗冲, 刘瑾. 基于深度学习的立体影像密集匹配方法综述[J]. 武汉大学学报 ( 信息科学版), 2021, 46(2): 193-202. DOI: 10.13203/j.whugis20200620
引用本文: 季顺平, 罗冲, 刘瑾. 基于深度学习的立体影像密集匹配方法综述[J]. 武汉大学学报 ( 信息科学版), 2021, 46(2): 193-202. DOI: 10.13203/j.whugis20200620
JI Shunping, LUO Chong, LIU Jin. A Review of Dense Stereo Image Matching Methods Based on Deep Learning[J]. Geomatics and Information Science of Wuhan University, 2021, 46(2): 193-202. DOI: 10.13203/j.whugis20200620
Citation: JI Shunping, LUO Chong, LIU Jin. A Review of Dense Stereo Image Matching Methods Based on Deep Learning[J]. Geomatics and Information Science of Wuhan University, 2021, 46(2): 193-202. DOI: 10.13203/j.whugis20200620

基于深度学习的立体影像密集匹配方法综述

A Review of Dense Stereo Image Matching Methods Based on Deep Learning

  • 摘要: 三维重建可用于数字高程模型制作、机器人导航、增强现实和自动驾驶等。视差图是三维重建中一种重要的表达方式,而立体密集匹配是使用最广泛的获取视差图的技术。近年来,随着硬件、数据集、算法的发展,基于深度学习的立体匹配方法受到了广泛关注并取得了巨大成功。然而,这些方法通常在近景立体像对中进行测试,很少被用于遥感影像中。回顾了双目立体匹配的深度学习方法,选出了代表性的5种经典深度学习模型——GC-Net(geometry and context network)模型、PSM-Net(pyramid stereo matching network)模型、GWC-Net(group-wise correlation stereo network)模型、GA-Net(guided aggregation network)模型、HSM-Net(hierarchical deep stereo matching network)模型,将其应用于一套开源街景数据集(KITTI2015)和两套航空遥感影像数据集(München、WHU);分析了各种网络的实现方法,探讨了深度学习在遥感影像立体匹配中的性能,并与传统方法进行了对比。

     

    Abstract: 3D reconstruction technology is widely used in digital elevation model production, robot navigation, augmented reality and autonomous driving, etc. Disparity map is an important expression of 3D reconstruction, and stereo matching is the most widely used technology to obtain a disparity map. In recent years, with the development of hardware, data sets, and algorithms, stereo matching methods based on deep learning have received extensive attention and achieved great success. However, these works are mainly validated in close-range images, and the evaluation on remote sensing aerial images is scarce. This paper reviews deep learning methods for stereo matching, and selects five representative models, such as GC-Net (geometry and context network), PSM-Net (pyramid stereo matching network), GWC-Net (group-wise correlation stereo network), GA-Net (guided aggregation network), HSM-Net (hierarchical deep stereo matching network), and applies them to a set of open source street-scene datasets (KITTI2015) and two sets of aerial remote sensing image datasets (München, WHU). The various networks are analyzed, and the performance of deep learning stereo matching methods is discussed and compared to traditional methods. The experimental results reveals that most of the deep learning methods exceed the classic semi-global matching and had a powerful generalization ability on cross-dataset transfer.

     

/

返回文章
返回