Correcting Pose Estimation with Implicit Occlusion Detection and Rectification

Ibrahim Hamed Ismail RADWAN, Abhinav Dhall, Roland Goecke

Research output: A Conference proceeding or a Chapter in BookConference contribution

1 Downloads (Pure)

Abstract

Recently, articulated pose estimation methods based on the pictorial structure framework have received much attention in computer vision. However, the performance of these approaches has been limited due to the presence of self-occlusion. This paper deals with the problem of handling self-occlusion in the pictorial structure framework. We propose an exemplar-based framework for implicit occlusion detection and rectification. Our framework can be applied as a general post-processing plug-in following any pose estimation approach to rectify errors due to self-occlusion and to improve the accuracy. The proposed framework outperforms a state-of-the-art pictorial structure approach for human pose estimation on the HumanEva dataset.
Original languageEnglish
Title of host publicationThe 2012 21th International Conference on Pattern Recognition (ICPR2012)
Place of PublicationTsukuba, Japan
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages3496-3499
Number of pages4
ISBN (Electronic)9784990644109
ISBN (Print)9781467322164
Publication statusPublished - 2012
Event21st International Conference on Pattern Recognition (ICPR 2012) - Tsukuba, Tsukuba, Japan
Duration: 11 Nov 201215 Nov 2012

Publication series

NameInternational Conference on Pattern Recognition

Conference

Conference21st International Conference on Pattern Recognition (ICPR 2012)
Abbreviated titleICPR 2012
CountryJapan
CityTsukuba
Period11/11/1215/11/12

Fingerprint

Computer vision
Processing

Cite this

RADWAN, I. H. I., Dhall, A., & Goecke, R. (2012). Correcting Pose Estimation with Implicit Occlusion Detection and Rectification. In The 2012 21th International Conference on Pattern Recognition (ICPR2012) (pp. 3496-3499). (International Conference on Pattern Recognition). Tsukuba, Japan: IEEE, Institute of Electrical and Electronics Engineers.
RADWAN, Ibrahim Hamed Ismail ; Dhall, Abhinav ; Goecke, Roland. / Correcting Pose Estimation with Implicit Occlusion Detection and Rectification. The 2012 21th International Conference on Pattern Recognition (ICPR2012). Tsukuba, Japan : IEEE, Institute of Electrical and Electronics Engineers, 2012. pp. 3496-3499 (International Conference on Pattern Recognition).
@inproceedings{37be36b7e0d943d0ba7c61d0a536d9f0,
title = "Correcting Pose Estimation with Implicit Occlusion Detection and Rectification",
abstract = "Recently, articulated pose estimation methods based on the pictorial structure framework have received much attention in computer vision. However, the performance of these approaches has been limited due to the presence of self-occlusion. This paper deals with the problem of handling self-occlusion in the pictorial structure framework. We propose an exemplar-based framework for implicit occlusion detection and rectification. Our framework can be applied as a general post-processing plug-in following any pose estimation approach to rectify errors due to self-occlusion and to improve the accuracy. The proposed framework outperforms a state-of-the-art pictorial structure approach for human pose estimation on the HumanEva dataset.",
keywords = "Human pose estimation, Occlusion detection",
author = "RADWAN, {Ibrahim Hamed Ismail} and Abhinav Dhall and Roland Goecke",
year = "2012",
language = "English",
isbn = "9781467322164",
series = "International Conference on Pattern Recognition",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
pages = "3496--3499",
booktitle = "The 2012 21th International Conference on Pattern Recognition (ICPR2012)",
address = "United States",

}

RADWAN, IHI, Dhall, A & Goecke, R 2012, Correcting Pose Estimation with Implicit Occlusion Detection and Rectification. in The 2012 21th International Conference on Pattern Recognition (ICPR2012). International Conference on Pattern Recognition, IEEE, Institute of Electrical and Electronics Engineers, Tsukuba, Japan, pp. 3496-3499, 21st International Conference on Pattern Recognition (ICPR 2012), Tsukuba, Japan, 11/11/12.

Correcting Pose Estimation with Implicit Occlusion Detection and Rectification. / RADWAN, Ibrahim Hamed Ismail; Dhall, Abhinav; Goecke, Roland.

The 2012 21th International Conference on Pattern Recognition (ICPR2012). Tsukuba, Japan : IEEE, Institute of Electrical and Electronics Engineers, 2012. p. 3496-3499 (International Conference on Pattern Recognition).

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Correcting Pose Estimation with Implicit Occlusion Detection and Rectification

AU - RADWAN, Ibrahim Hamed Ismail

AU - Dhall, Abhinav

AU - Goecke, Roland

PY - 2012

Y1 - 2012

N2 - Recently, articulated pose estimation methods based on the pictorial structure framework have received much attention in computer vision. However, the performance of these approaches has been limited due to the presence of self-occlusion. This paper deals with the problem of handling self-occlusion in the pictorial structure framework. We propose an exemplar-based framework for implicit occlusion detection and rectification. Our framework can be applied as a general post-processing plug-in following any pose estimation approach to rectify errors due to self-occlusion and to improve the accuracy. The proposed framework outperforms a state-of-the-art pictorial structure approach for human pose estimation on the HumanEva dataset.

AB - Recently, articulated pose estimation methods based on the pictorial structure framework have received much attention in computer vision. However, the performance of these approaches has been limited due to the presence of self-occlusion. This paper deals with the problem of handling self-occlusion in the pictorial structure framework. We propose an exemplar-based framework for implicit occlusion detection and rectification. Our framework can be applied as a general post-processing plug-in following any pose estimation approach to rectify errors due to self-occlusion and to improve the accuracy. The proposed framework outperforms a state-of-the-art pictorial structure approach for human pose estimation on the HumanEva dataset.

KW - Human pose estimation

KW - Occlusion detection

M3 - Conference contribution

SN - 9781467322164

T3 - International Conference on Pattern Recognition

SP - 3496

EP - 3499

BT - The 2012 21th International Conference on Pattern Recognition (ICPR2012)

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Tsukuba, Japan

ER -

RADWAN IHI, Dhall A, Goecke R. Correcting Pose Estimation with Implicit Occlusion Detection and Rectification. In The 2012 21th International Conference on Pattern Recognition (ICPR2012). Tsukuba, Japan: IEEE, Institute of Electrical and Electronics Engineers. 2012. p. 3496-3499. (International Conference on Pattern Recognition).