Bilateral Grid Learning for Stereo Matching Networks
This work provides a method for improving the real-time performance of stereo matching networks for applications like autonomous driving and robotics, representing an incremental improvement in the field.
This paper addresses the challenge of balancing real-time performance and accuracy in stereo matching networks by introducing a novel edge-preserving cost volume upsampling module. This module, based on a learned bilateral grid, accelerates existing networks several times while maintaining comparable accuracy, and enables a new network (BGNet) to outperform existing real-time deep stereo matching networks on KITTI datasets.
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 real-time 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.