In this paper, a new deep learning architecture for stereo disparity
estimation is proposed. The proposed atrous multiscale network (AMNet) adopts
an efficient feature extractor with depthwise-separable convolutions and an
extended cost volume that deploys novel stereo matching costs on the deep
features. A stacked atrous multiscale network is proposed to aggregate rich
multiscale contextual information from the cost volume which allows for
estimating the disparity with high accuracy at multiple scales. AMNet can be
further modified to be a foreground-background aware network, FBA-AMNet, which
is capable of discriminating between the foreground and the background objects
in the scene at multiple scales. An iterative multitask learning method is
proposed to train FBA-AMNet end-to-end. The proposed disparity estimation
networks, AMNet and FBA-AMNet, show accurate disparity estimates and advance
the state of the art on the challenging Middlebury, KITTI 2012, KITTI 2015, and
Sceneflow stereo disparity estimation benchmarks.

本文提出了一种新的深度学习架构 ——Atrous Multiscale Network（AMNet）用于立体视差估计，AMNet 采用了有效的特征提取器，并且能够在深度特征上部署新的立体匹配代价，其提出了堆叠式 Atrous Multiscale 网络来聚合来自代价体积的丰富多尺度语境信息，可在多种尺度上高精度地估计视差。此外，本文提出了一种迭代式多任务学习方法来训练前景 - 背景感知 Atrous Multiscale 网络（FBA-AMNet），并且表明 AMNet 和 FBA-AMNet 在具有挑战性的 Middlebury，KITTI 2012，KITTI 2015 和 Sceneflow 立体视差估计基准测试中具有高精度的视差估计。