Depth recovery via decomposition of polynomial and piece-wise constant signals

Xinchen Ye, Xiaolin Song, Jingyu Yang, Chunping Hou, Yao Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper proposes a novel decomposition model for high-quality depth recovery (DMDR) from low quality depth measurement accompanied by high-resolution RGB image. We observe that depth patches extracted from the depth map containing smooth regions separated by curves, can be decomposed simultaneously by a low-order polynomial surface and a piece-wise constant signal. In our model, the polynomial surface component is regularized by least-square polynomial smoothing, while the piece-wise constant component is constrained by total variation filtering. The model is effectively solved by the alternating direction method under the augmented Lagrangian multiplier (ALM-ADM) algorithm. Experimental results show that our method is able to handle various types of depth degradation under the designed signal decomposition model, and produces high-quality depth recovery results.

Original languageEnglish (US)
Title of host publicationVCIP 2016 - 30th Anniversary of Visual Communication and Image Processing
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509053162
DOIs
StatePublished - Jan 4 2017
Event2016 IEEE Visual Communication and Image Processing, VCIP 2016 - Chengdu, China
Duration: Nov 27 2016Nov 30 2016

Other

Other2016 IEEE Visual Communication and Image Processing, VCIP 2016
CountryChina
CityChengdu
Period11/27/1611/30/16

Fingerprint

Polynomials
Decomposition
Recovery
Delta modulation
Image resolution
Degradation

Keywords

  • decomposition
  • Depth recovery
  • piece-wise constant
  • polynomial
  • Total variation

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Ye, X., Song, X., Yang, J., Hou, C., & Wang, Y. (2017). Depth recovery via decomposition of polynomial and piece-wise constant signals. In VCIP 2016 - 30th Anniversary of Visual Communication and Image Processing [7805434] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VCIP.2016.7805434

Depth recovery via decomposition of polynomial and piece-wise constant signals. / Ye, Xinchen; Song, Xiaolin; Yang, Jingyu; Hou, Chunping; Wang, Yao.

VCIP 2016 - 30th Anniversary of Visual Communication and Image Processing. Institute of Electrical and Electronics Engineers Inc., 2017. 7805434.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ye, X, Song, X, Yang, J, Hou, C & Wang, Y 2017, Depth recovery via decomposition of polynomial and piece-wise constant signals. in VCIP 2016 - 30th Anniversary of Visual Communication and Image Processing., 7805434, Institute of Electrical and Electronics Engineers Inc., 2016 IEEE Visual Communication and Image Processing, VCIP 2016, Chengdu, China, 11/27/16. https://doi.org/10.1109/VCIP.2016.7805434
Ye X, Song X, Yang J, Hou C, Wang Y. Depth recovery via decomposition of polynomial and piece-wise constant signals. In VCIP 2016 - 30th Anniversary of Visual Communication and Image Processing. Institute of Electrical and Electronics Engineers Inc. 2017. 7805434 https://doi.org/10.1109/VCIP.2016.7805434
Ye, Xinchen ; Song, Xiaolin ; Yang, Jingyu ; Hou, Chunping ; Wang, Yao. / Depth recovery via decomposition of polynomial and piece-wise constant signals. VCIP 2016 - 30th Anniversary of Visual Communication and Image Processing. Institute of Electrical and Electronics Engineers Inc., 2017.
@inproceedings{91213f2d6c344d1ea6c62df5158774ac,
title = "Depth recovery via decomposition of polynomial and piece-wise constant signals",
abstract = "This paper proposes a novel decomposition model for high-quality depth recovery (DMDR) from low quality depth measurement accompanied by high-resolution RGB image. We observe that depth patches extracted from the depth map containing smooth regions separated by curves, can be decomposed simultaneously by a low-order polynomial surface and a piece-wise constant signal. In our model, the polynomial surface component is regularized by least-square polynomial smoothing, while the piece-wise constant component is constrained by total variation filtering. The model is effectively solved by the alternating direction method under the augmented Lagrangian multiplier (ALM-ADM) algorithm. Experimental results show that our method is able to handle various types of depth degradation under the designed signal decomposition model, and produces high-quality depth recovery results.",
keywords = "decomposition, Depth recovery, piece-wise constant, polynomial, Total variation",
author = "Xinchen Ye and Xiaolin Song and Jingyu Yang and Chunping Hou and Yao Wang",
year = "2017",
month = "1",
day = "4",
doi = "10.1109/VCIP.2016.7805434",
language = "English (US)",
booktitle = "VCIP 2016 - 30th Anniversary of Visual Communication and Image Processing",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Depth recovery via decomposition of polynomial and piece-wise constant signals

AU - Ye, Xinchen

AU - Song, Xiaolin

AU - Yang, Jingyu

AU - Hou, Chunping

AU - Wang, Yao

PY - 2017/1/4

Y1 - 2017/1/4

N2 - This paper proposes a novel decomposition model for high-quality depth recovery (DMDR) from low quality depth measurement accompanied by high-resolution RGB image. We observe that depth patches extracted from the depth map containing smooth regions separated by curves, can be decomposed simultaneously by a low-order polynomial surface and a piece-wise constant signal. In our model, the polynomial surface component is regularized by least-square polynomial smoothing, while the piece-wise constant component is constrained by total variation filtering. The model is effectively solved by the alternating direction method under the augmented Lagrangian multiplier (ALM-ADM) algorithm. Experimental results show that our method is able to handle various types of depth degradation under the designed signal decomposition model, and produces high-quality depth recovery results.

AB - This paper proposes a novel decomposition model for high-quality depth recovery (DMDR) from low quality depth measurement accompanied by high-resolution RGB image. We observe that depth patches extracted from the depth map containing smooth regions separated by curves, can be decomposed simultaneously by a low-order polynomial surface and a piece-wise constant signal. In our model, the polynomial surface component is regularized by least-square polynomial smoothing, while the piece-wise constant component is constrained by total variation filtering. The model is effectively solved by the alternating direction method under the augmented Lagrangian multiplier (ALM-ADM) algorithm. Experimental results show that our method is able to handle various types of depth degradation under the designed signal decomposition model, and produces high-quality depth recovery results.

KW - decomposition

KW - Depth recovery

KW - piece-wise constant

KW - polynomial

KW - Total variation

UR - http://www.scopus.com/inward/record.url?scp=85011072594&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011072594&partnerID=8YFLogxK

U2 - 10.1109/VCIP.2016.7805434

DO - 10.1109/VCIP.2016.7805434

M3 - Conference contribution

AN - SCOPUS:85011072594

BT - VCIP 2016 - 30th Anniversary of Visual Communication and Image Processing

PB - Institute of Electrical and Electronics Engineers Inc.

ER -