Journal
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
Volume -, Issue -, Pages 12492-12501Publisher
IEEE COMPUTER SOC
DOI: 10.1109/CVPR46437.2021.01231
Keywords
-
Funding
- Orbbec Inc. [W2020JSKF0547]
- National Natural Science Foundation of China [U20A20185, 61972435, 62076086]
- Natural Science Foundation of Guangdong Province [2019A1515011271]
- Shenzhen Science and Technology Program [RCYX20200714114641140, JCYJ20190807152209394]
Ask authors/readers for more resources
This paper introduces a novel module for stereo matching networks that improves real-time performance and maintains accuracy. The module can be seamlessly embedded into existing networks to achieve significant acceleration. The real-time network based on this module outperforms existing networks on the KITTI dataset.
Real-time performance of stereo matching networks is important for many applications, such as automatic driving, robot navigation and augmented reality (AR). Although significant progress has been made in stereo matching networks in recent years, it is still challenging to balance realtime performance and accuracy. In this paper, we present a novel edge-preserving cost volume upsampling module based on the slicing operation in the learned bilateral grid. The slicing layer is parameter-free, which allows us to obtain a high quality cost volume of high resolution from a low-resolution cost volume under the guide of the learned guidance map efficiently. The proposed cost volume upsampling module can be seamlessly embedded into many existing stereo matching networks, such as GCNet, PSMNet, and GANet. The resulting networks are accelerated several times while maintaining comparable accuracy. Furthermore, we design a real-time network (named BGNet) based on this module, which outperforms existing published real-time deep stereo matching networks, as well as some complex networks on the KITTI stereo datasets. The code is available at https://github.com/YuhuaXu/BGNet.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available