DeepMVS: Learning Multi-view Stereopsis

We present DeepMVS, a deep convolutional neural network (ConvNet) for multi-view stereo reconstruction. Taking an arbitrary number of posed images as input, we first produce a set of plane-sweep volumes and use the proposed DeepMVS network to predict high-quality disparity maps. The key contribution...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 2821 - 2830
Main Authors Huang, Po-Han, Matzen, Kevin, Kopf, Johannes, Ahuja, Narendra, Huang, Jia-Bin
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2018
Subjects
Calibration
Cameras
Decoding
Feature extraction
Image reconstruction
Semantics
Training
Online AccessGet full text

Cover

More Information
Summary:We present DeepMVS, a deep convolutional neural network (ConvNet) for multi-view stereo reconstruction. Taking an arbitrary number of posed images as input, we first produce a set of plane-sweep volumes and use the proposed DeepMVS network to predict high-quality disparity maps. The key contributions that enable these results are (1) supervised pretraining on a photorealistic synthetic dataset, (2) an effective method for aggregating information across a set of unordered images, and (3) integrating multi-layer feature activations from the pre-trained VGG-19 network. We validate the efficacy of DeepMVS using the ETH3D Benchmark. Our results show that DeepMVS compares favorably against state-of-the-art conventional MVS algorithms and other ConvNet based methods, particularly for near-textureless regions and thin structures.
ISSN:1063-6919
DOI:10.1109/CVPR.2018.00298