Discriminative Middle-Level Parts Mining for Object Detection

Dong LI; Yali LI; Shengjin WANG

doi:10.1587/transinf.2015EDP7200

Discriminative Middle-Level Parts Mining for Object Detection

Dong LI, Yali LI, Shengjin WANG

Full Text Views

0

Share
Cite this

Summary :

Middle-level parts have attracted great attention in the computer vision community, acting as discriminative elements for objects. In this paper we propose an unsupervised approach to mine discriminative parts for object detection. This work features three aspects. First, we introduce an unsupervised, exemplar-based training process for part detection. We generate initial parts by selective search and then train part detectors by exemplar SVM. Second, a part selection model based on consistency and distinctiveness is constructed to select effective parts from the candidate pool. Third, we combine discriminative part mining with the deformable part model (DPM) for object detection. The proposed method is evaluated on the PASCAL VOC2007 and VOC2010 datasets. The experimental results demons-trate the effectiveness of our method for object detection.

Publication: IEICE TRANSACTIONS on Information Vol.E98-D No.11 pp.1950-1957

Publication Date: 2015/11/01

Publicized: 2015/08/03

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2015EDP7200

Type of Manuscript: PAPER

Category: Image Recognition, Computer Vision

Authors

Dong LI
  Tsinghua University
Yali LI
  Tsinghua University
Shengjin WANG
  Tsinghua University

Keyword

object detection, discriminative part mining, unsupervised part training, part selection

Cite this

Copy

Dong LI, Yali LI, Shengjin WANG, "Discriminative Middle-Level Parts Mining for Object Detection" in IEICE TRANSACTIONS on Information, vol. E98-D, no. 11, pp. 1950-1957, November 2015, doi: 10.1587/transinf.2015EDP7200.
Abstract: Middle-level parts have attracted great attention in the computer vision community, acting as discriminative elements for objects. In this paper we propose an unsupervised approach to mine discriminative parts for object detection. This work features three aspects. First, we introduce an unsupervised, exemplar-based training process for part detection. We generate initial parts by selective search and then train part detectors by exemplar SVM. Second, a part selection model based on consistency and distinctiveness is constructed to select effective parts from the candidate pool. Third, we combine discriminative part mining with the deformable part model (DPM) for object detection. The proposed method is evaluated on the PASCAL VOC2007 and VOC2010 datasets. The experimental results demons-trate the effectiveness of our method for object detection.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.2015EDP7200/_p

Copy

@ARTICLE{e98-d_11_1950,
author={Dong LI, Yali LI, Shengjin WANG, },
journal={IEICE TRANSACTIONS on Information},
title={Discriminative Middle-Level Parts Mining for Object Detection},
year={2015},
volume={E98-D},
number={11},
pages={1950-1957},
abstract={Middle-level parts have attracted great attention in the computer vision community, acting as discriminative elements for objects. In this paper we propose an unsupervised approach to mine discriminative parts for object detection. This work features three aspects. First, we introduce an unsupervised, exemplar-based training process for part detection. We generate initial parts by selective search and then train part detectors by exemplar SVM. Second, a part selection model based on consistency and distinctiveness is constructed to select effective parts from the candidate pool. Third, we combine discriminative part mining with the deformable part model (DPM) for object detection. The proposed method is evaluated on the PASCAL VOC2007 and VOC2010 datasets. The experimental results demons-trate the effectiveness of our method for object detection.},
keywords={},
doi={10.1587/transinf.2015EDP7200},
ISSN={1745-1361},
month={November},}

Copy

TY - JOUR
TI - Discriminative Middle-Level Parts Mining for Object Detection
T2 - IEICE TRANSACTIONS on Information
SP - 1950
EP - 1957
AU - Dong LI
AU - Yali LI
AU - Shengjin WANG
PY - 2015
DO - 10.1587/transinf.2015EDP7200
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2015
AB - Middle-level parts have attracted great attention in the computer vision community, acting as discriminative elements for objects. In this paper we propose an unsupervised approach to mine discriminative parts for object detection. This work features three aspects. First, we introduce an unsupervised, exemplar-based training process for part detection. We generate initial parts by selective search and then train part detectors by exemplar SVM. Second, a part selection model based on consistency and distinctiveness is constructed to select effective parts from the candidate pool. Third, we combine discriminative part mining with the deformable part model (DPM) for object detection. The proposed method is evaluated on the PASCAL VOC2007 and VOC2010 datasets. The experimental results demons-trate the effectiveness of our method for object detection.
ER -