HMM-based surface reconstruction from single images

Takayuki Nagai, Masaaki Ikehara, Akira Kurematsu

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

In this paper, a novel method of surface reconstruction from a single monocular image is proposed. Our proposed approach (called "shape from object-specific knowledge") is based on knowledge of objects, acquired by learning from a number of samples. To achieve this, we investigate the use of the Subband Pseudo 2D Hidden Markov Model (SPHMM), which is an extended version of normal Pseudo 2D HMMs. SPHMMs can model the correspondence between an intensity image and its depth information. We have applied our algorithm to 3D face, 3D hand, and 3D car reconstruction from single images, and the results show the effectiveness of the proposed method.

Original languageEnglish
Pages (from-to)80-89
Number of pages10
JournalSystems and Computers in Japan
Volume38
Issue number11
DOIs
Publication statusPublished - 2007 Oct

Fingerprint

Surface Reconstruction
Surface reconstruction
Hidden Markov models
Railroad cars
Markov Model
Correspondence
Face
Object
Knowledge
Model

Keywords

  • 3D shape recovery
  • HMM
  • Object recognition
  • Object-specific knowledge

ASJC Scopus subject areas

  • Hardware and Architecture
  • Information Systems
  • Theoretical Computer Science
  • Computational Theory and Mathematics

Cite this

HMM-based surface reconstruction from single images. / Nagai, Takayuki; Ikehara, Masaaki; Kurematsu, Akira.

In: Systems and Computers in Japan, Vol. 38, No. 11, 10.2007, p. 80-89.

Research output: Contribution to journalArticle

Nagai, Takayuki ; Ikehara, Masaaki ; Kurematsu, Akira. / HMM-based surface reconstruction from single images. In: Systems and Computers in Japan. 2007 ; Vol. 38, No. 11. pp. 80-89.
@article{ac3d9818e0b94cbdbce485cde34c1d4e,
title = "HMM-based surface reconstruction from single images",
abstract = "In this paper, a novel method of surface reconstruction from a single monocular image is proposed. Our proposed approach (called {"}shape from object-specific knowledge{"}) is based on knowledge of objects, acquired by learning from a number of samples. To achieve this, we investigate the use of the Subband Pseudo 2D Hidden Markov Model (SPHMM), which is an extended version of normal Pseudo 2D HMMs. SPHMMs can model the correspondence between an intensity image and its depth information. We have applied our algorithm to 3D face, 3D hand, and 3D car reconstruction from single images, and the results show the effectiveness of the proposed method.",
keywords = "3D shape recovery, HMM, Object recognition, Object-specific knowledge",
author = "Takayuki Nagai and Masaaki Ikehara and Akira Kurematsu",
year = "2007",
month = "10",
doi = "10.1002/scj.10685",
language = "English",
volume = "38",
pages = "80--89",
journal = "Systems and Computers in Japan",
issn = "0882-1666",
publisher = "John Wiley and Sons Inc.",
number = "11",

}

TY - JOUR

T1 - HMM-based surface reconstruction from single images

AU - Nagai, Takayuki

AU - Ikehara, Masaaki

AU - Kurematsu, Akira

PY - 2007/10

Y1 - 2007/10

N2 - In this paper, a novel method of surface reconstruction from a single monocular image is proposed. Our proposed approach (called "shape from object-specific knowledge") is based on knowledge of objects, acquired by learning from a number of samples. To achieve this, we investigate the use of the Subband Pseudo 2D Hidden Markov Model (SPHMM), which is an extended version of normal Pseudo 2D HMMs. SPHMMs can model the correspondence between an intensity image and its depth information. We have applied our algorithm to 3D face, 3D hand, and 3D car reconstruction from single images, and the results show the effectiveness of the proposed method.

AB - In this paper, a novel method of surface reconstruction from a single monocular image is proposed. Our proposed approach (called "shape from object-specific knowledge") is based on knowledge of objects, acquired by learning from a number of samples. To achieve this, we investigate the use of the Subband Pseudo 2D Hidden Markov Model (SPHMM), which is an extended version of normal Pseudo 2D HMMs. SPHMMs can model the correspondence between an intensity image and its depth information. We have applied our algorithm to 3D face, 3D hand, and 3D car reconstruction from single images, and the results show the effectiveness of the proposed method.

KW - 3D shape recovery

KW - HMM

KW - Object recognition

KW - Object-specific knowledge

UR - http://www.scopus.com/inward/record.url?scp=34547985743&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547985743&partnerID=8YFLogxK

U2 - 10.1002/scj.10685

DO - 10.1002/scj.10685

M3 - Article

AN - SCOPUS:34547985743

VL - 38

SP - 80

EP - 89

JO - Systems and Computers in Japan

JF - Systems and Computers in Japan

SN - 0882-1666

IS - 11

ER -