Title

High Quality 3D Reconstruction from Multiple Images

Alternative Title

다중 이미지를 이용한 고품질 3D 복원 방법

DGIST Authors

Kyumin Hwang ; Sunghoon Im ; Kyong Hwan Jin

Advisor

임성훈

Co-Advisor(s)

Kyong Hwan Jin

Issued Date

2022

Awarded Date

2022/02

Citation

Kyumin Hwang. (2022). High Quality 3D Reconstruction from Multiple Images. doi: 10.22677/thesis.200000593272

Type

Thesis

Subject

Multi-view Stereo (MVS), Depth Estimation, Normal Estimation

Description

Multi-view Stereo (MVS), Depth Estimation, Normal Estimation

Table Of Contents

I. Introduction 1
II. Related Work 3
III. Proposed method 4
3.1 Initial Depth and Normal Estimation 4
3.2 Cost Volume Warping & Fusion 6
3.3 Geometric Consistency-based Refinement 7
IV. Experiments 9
4.1 Evaluation on Depth Estimation 12
4.2 Evaluation on Normal Estimation 14
V. Ablation Study 15
5.1 Ablation Study on Consistency Map Representation Method 15
5.2 Ablation Study on Multi-view Matching 16
5.3 Ablation Study on Multi-directional Plane Sweep Algorithm 16
5.4 Ablation Study on Novel View Synthesis Application 18
VI. Conclusion 24
VII. References 25
VIII. 요약문 29

URI

http://dgist.dcollection.net/common/orgView/200000593272
http://hdl.handle.net/20.500.11750/16287

DOI

10.22677/thesis.200000593272

Degree

Master

Department

Information and Communication Engineering

Publisher

DGIST

Show Full Item Record