Octave Deep Plane-Sweeping Network: Reducing Spatial Redundancy for Learning-Based Plane-Sweeping Stereo

Octave Deep Plane-Sweeping Network: Reducing Spatial Redundancy for Learning-Based Plane-Sweeping Stereo

In this paper, we propose the octave deep plane-sweeping network (OctDPSNet). OctDPSNet is a novel learning-based plane-sweeping stereo, which drastically reduces the required GPU memory and computation time while achieving a state-of-the-art depth estimation accuracy. Inspired by octave convolution...

Full description

Bibliographic Details
Main Authors: Ren Komatsu, Hiromitsu Fujii, Yusuke Tamura, Atsushi Yamashita, Hajime Asama
Format: Article
Language:English
Published: IEEE 2019-01-01
Series:IEEE Access
Subjects:
Convolutional neural networks
deep neural networks
depth reconstruction
plane-sweeping stereo
stereo vision
Online Access:https://ieeexplore.ieee.org/document/8867874/

Internet

https://ieeexplore.ieee.org/document/8867874/

Similar Items