Files in this item

FilesDescriptionFormat

application/pdf

application/pdfHUANG-THESIS-2018.pdf (11MB)
(no description provided)PDF

Description

Title:DeepMVS: learning multi-view stereopsis
Author(s):Huang, Po-Han
Advisor(s):Ahuja, Narendra
Department / Program:Electrical & Computer Eng
Discipline:Electrical & Computer Engr
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:M.S.
Genre:Thesis
Subject(s):computer vision, deep learning, multi-view stereo, DeepMVS, pattern recognition, 3D reconstruction, depth estimation, machine learning, convolutional neural network
Abstract: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.
Issue Date:2018-06-25
Type:Text
URI:http://hdl.handle.net/2142/101770
Rights Information:Copyright 2018 Po-Han Huang
Date Available in IDEALS:2018-09-27
2020-09-28
Date Deposited:2018-08


This item appears in the following Collection(s)

Item Statistics