Depth superresolution by transduction

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

This paper presents a depth superresolution (SR) method that uses both of a low-resolution (LR) depth image and a high-resolution (HR) intensity image. We formulate depth SR as a graph-based transduction problem. In particular, the HR intensity image is represented as an undirected graph, in which pixels are characterized as vertices, and their relations are encoded as an affinity function. When the vertices initially labeled with certain depth hypotheses (from the LR depth image) are regarded as input queries, all the vertices are scored with respect to the relevances to these queries by a classifying function. Each vertex is then labeled with the depth hypothesis that receives the highest relevance score. We design the classifying function by considering the local and global structures of the HR intensity image. This approach enables us to address a depth bleeding problem that typically appears in current depth SR methods. Furthermore, input queries are assigned in a probabilistic manner, making depth SR robust to noisy depth measurements. We also analyze existing depth SR methods in the context of transduction, and discuss their theoretic relations. Intensive experiments demonstrate the superiority of the proposed method over state-of-the-art methods both qualitatively and quantitatively.

Original languageEnglish
Article number7047894
Pages (from-to)1524-1535
Number of pages12
JournalIEEE Transactions on Image Processing
Volume24
Issue number5
DOIs
Publication statusPublished - 2015 May 1

Fingerprint

Pixels
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

@article{f26b168917a34d819c4373a42925161e,
title = "Depth superresolution by transduction",
abstract = "This paper presents a depth superresolution (SR) method that uses both of a low-resolution (LR) depth image and a high-resolution (HR) intensity image. We formulate depth SR as a graph-based transduction problem. In particular, the HR intensity image is represented as an undirected graph, in which pixels are characterized as vertices, and their relations are encoded as an affinity function. When the vertices initially labeled with certain depth hypotheses (from the LR depth image) are regarded as input queries, all the vertices are scored with respect to the relevances to these queries by a classifying function. Each vertex is then labeled with the depth hypothesis that receives the highest relevance score. We design the classifying function by considering the local and global structures of the HR intensity image. This approach enables us to address a depth bleeding problem that typically appears in current depth SR methods. Furthermore, input queries are assigned in a probabilistic manner, making depth SR robust to noisy depth measurements. We also analyze existing depth SR methods in the context of transduction, and discuss their theoretic relations. Intensive experiments demonstrate the superiority of the proposed method over state-of-the-art methods both qualitatively and quantitatively.",
author = "Bumsub Ham and Dongbo Min and Kwanghoon Sohn",
year = "2015",
month = "5",
day = "1",
doi = "10.1109/TIP.2015.2405342",
language = "English",
volume = "24",
pages = "1524--1535",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

Depth superresolution by transduction. / Ham, Bumsub; Min, Dongbo; Sohn, Kwanghoon.

In: IEEE Transactions on Image Processing, Vol. 24, No. 5, 7047894, 01.05.2015, p. 1524-1535.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Depth superresolution by transduction

AU - Ham, Bumsub

AU - Min, Dongbo

AU - Sohn, Kwanghoon

PY - 2015/5/1

Y1 - 2015/5/1

N2 - This paper presents a depth superresolution (SR) method that uses both of a low-resolution (LR) depth image and a high-resolution (HR) intensity image. We formulate depth SR as a graph-based transduction problem. In particular, the HR intensity image is represented as an undirected graph, in which pixels are characterized as vertices, and their relations are encoded as an affinity function. When the vertices initially labeled with certain depth hypotheses (from the LR depth image) are regarded as input queries, all the vertices are scored with respect to the relevances to these queries by a classifying function. Each vertex is then labeled with the depth hypothesis that receives the highest relevance score. We design the classifying function by considering the local and global structures of the HR intensity image. This approach enables us to address a depth bleeding problem that typically appears in current depth SR methods. Furthermore, input queries are assigned in a probabilistic manner, making depth SR robust to noisy depth measurements. We also analyze existing depth SR methods in the context of transduction, and discuss their theoretic relations. Intensive experiments demonstrate the superiority of the proposed method over state-of-the-art methods both qualitatively and quantitatively.

AB - This paper presents a depth superresolution (SR) method that uses both of a low-resolution (LR) depth image and a high-resolution (HR) intensity image. We formulate depth SR as a graph-based transduction problem. In particular, the HR intensity image is represented as an undirected graph, in which pixels are characterized as vertices, and their relations are encoded as an affinity function. When the vertices initially labeled with certain depth hypotheses (from the LR depth image) are regarded as input queries, all the vertices are scored with respect to the relevances to these queries by a classifying function. Each vertex is then labeled with the depth hypothesis that receives the highest relevance score. We design the classifying function by considering the local and global structures of the HR intensity image. This approach enables us to address a depth bleeding problem that typically appears in current depth SR methods. Furthermore, input queries are assigned in a probabilistic manner, making depth SR robust to noisy depth measurements. We also analyze existing depth SR methods in the context of transduction, and discuss their theoretic relations. Intensive experiments demonstrate the superiority of the proposed method over state-of-the-art methods both qualitatively and quantitatively.

UR - http://www.scopus.com/inward/record.url?scp=84925240129&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84925240129&partnerID=8YFLogxK

U2 - 10.1109/TIP.2015.2405342

DO - 10.1109/TIP.2015.2405342

M3 - Article

AN - SCOPUS:84925240129

VL - 24

SP - 1524

EP - 1535

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 5

M1 - 7047894

ER -