The KITTI Vision Benchmark Suite

Deep neural networks significantly enhance the
accuracy of the stereo-based depth measurement.
Some current methods suffer from inefficient use
of the global context information, which will lead
to the loss of structural details in ill-posed
areas. To this end, we design a novel stereo
network GAMNet composed of three core components
(GDA, MPF, DCA) for measuring the depth
information in challenging real-world
environments. Firstly, we present a lightweight
attention module, integrating the global semantic
cues for every feature position across the channel
and spatial dimensions. Next, we construct the MPF
module to fuse the diverse semantic and contextual
information from different levels of the feature
pyramid. Finally, we aggregate cost volume with a
stacked encoder-decoder composed of our DCA module
and 3D convolutions, filtering the transmission of
matching clues and capturing the rich global
contexts.

Number of model parameters: 7125754