Accurate depth inference of multi-view stereo using deep neural network심층 신경망을 활용한 다중 시점 스테레오의 정확한 깊이값 추정

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 531
  • Download : 0
We propose an accurate depth inference of multi-view stereo using deep neural network, which aims to compute accurate depths on both fine structures and overall scene. Inspired by learning-based approach of plane sweeping algorithm, we design Deep Neural Networks (DNN) targeting both preserving sharp boundaries and inferring reasonable depths even in homogeneous texture regions. Our main contributions to achieve the goal is design of separate patch matching network depending on their tasks by arranging local and global patch matching in parallel, and also reducing the memory consumption and inference time into half while maintaining the performance. We trained our network using well-known MVS benchmarks, and validated our performance on ETH3D benchmark, which revealed that ours showed more accurate depth prediction compared to state-of-the-art learning-based multi-view stereo algorithms. As a result, our system enables high-quality depth predictions, which possibly leads to denser and more precise 3D dense reconstruction.
Advisors
Woo, Woontackresearcher우운택researcher
Description
한국과학기술원 :문화기술대학원,
Publisher
한국과학기술원
Issue Date
2019
Identifier
325007
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 문화기술대학원, 2019.2,[iii, 32 p. :]

Keywords

computer vision▼amulti-view stereo▼adeep neural network▼adepth estimation▼apatch matching; 컴퓨터 비전▼a다중 시점 스테레오▼a심층 신경망▼a깊이값 추정▼a패치 매칭

URI
http://hdl.handle.net/10203/266033
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=843084&flag=dissertation
Appears in Collection
GCT-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0