Discriminative Semantic Parts Learning for Object Detection

Yurui XIE; Qingbo WU; Bing LUO

doi:10.1587/transinf.2015EDL8014

Discriminative Semantic Parts Learning for Object Detection

Yurui XIE, Qingbo WU, Bing LUO

Full Text Views

0

Cite this

Summary :

In this letter, we propose a new semantic parts learning approach to address the object detection problem with only the bounding boxes of object category labels. Our main observation is that even though the appearance and arrangement of object parts might have variations across the instances of different object categories, the constituent parts still maintain geometric consistency. Specifically, we propose a discriminative clustering method with sparse representation refinement to discover the mid-level semantic part set automatically. Then each semantic part detector is learned by the linear SVM in a one-vs-all manner. Finally, we utilize the learned part detectors to score the test image and integrate all the response maps of part detectors to obtain the detection result. The learned class-generic part detectors have the ability to capture the objects across different categories. Experimental results show that the performance of our approach can outperform some recent competing methods.

Publication: IEICE TRANSACTIONS on Information Vol.E98-D No.7 pp.1434-1438

Publication Date: 2015/07/01

Publicized: 2015/04/15

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2015EDL8014

Type of Manuscript: LETTER

Category: Image Recognition, Computer Vision

Authors

Yurui XIE
  University of Electronic Science and Technology of China
Qingbo WU
  University of Electronic Science and Technology of China
Bing LUO
  University of Electronic Science and Technology of China

Keyword

object detection, sparse representation

Cite this

Copy

Yurui XIE, Qingbo WU, Bing LUO, "Discriminative Semantic Parts Learning for Object Detection" in IEICE TRANSACTIONS on Information, vol. E98-D, no. 7, pp. 1434-1438, July 2015, doi: 10.1587/transinf.2015EDL8014.
Abstract: In this letter, we propose a new semantic parts learning approach to address the object detection problem with only the bounding boxes of object category labels. Our main observation is that even though the appearance and arrangement of object parts might have variations across the instances of different object categories, the constituent parts still maintain geometric consistency. Specifically, we propose a discriminative clustering method with sparse representation refinement to discover the mid-level semantic part set automatically. Then each semantic part detector is learned by the linear SVM in a one-vs-all manner. Finally, we utilize the learned part detectors to score the test image and integrate all the response maps of part detectors to obtain the detection result. The learned class-generic part detectors have the ability to capture the objects across different categories. Experimental results show that the performance of our approach can outperform some recent competing methods.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2015EDL8014/_p

Copy

@ARTICLE{e98-d_7_1434,
author={Yurui XIE, Qingbo WU, Bing LUO, },
journal={IEICE TRANSACTIONS on Information},
title={Discriminative Semantic Parts Learning for Object Detection},
year={2015},
volume={E98-D},
number={7},
pages={1434-1438},
abstract={In this letter, we propose a new semantic parts learning approach to address the object detection problem with only the bounding boxes of object category labels. Our main observation is that even though the appearance and arrangement of object parts might have variations across the instances of different object categories, the constituent parts still maintain geometric consistency. Specifically, we propose a discriminative clustering method with sparse representation refinement to discover the mid-level semantic part set automatically. Then each semantic part detector is learned by the linear SVM in a one-vs-all manner. Finally, we utilize the learned part detectors to score the test image and integrate all the response maps of part detectors to obtain the detection result. The learned class-generic part detectors have the ability to capture the objects across different categories. Experimental results show that the performance of our approach can outperform some recent competing methods.},
keywords={},
doi={10.1587/transinf.2015EDL8014},
ISSN={1745-1361},
month={July},}

Copy

TY - JOUR
TI - Discriminative Semantic Parts Learning for Object Detection
T2 - IEICE TRANSACTIONS on Information
SP - 1434
EP - 1438
AU - Yurui XIE
AU - Qingbo WU
AU - Bing LUO
PY - 2015
DO - 10.1587/transinf.2015EDL8014
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2015
AB - In this letter, we propose a new semantic parts learning approach to address the object detection problem with only the bounding boxes of object category labels. Our main observation is that even though the appearance and arrangement of object parts might have variations across the instances of different object categories, the constituent parts still maintain geometric consistency. Specifically, we propose a discriminative clustering method with sparse representation refinement to discover the mid-level semantic part set automatically. Then each semantic part detector is learned by the linear SVM in a one-vs-all manner. Finally, we utilize the learned part detectors to score the test image and integrate all the response maps of part detectors to obtain the detection result. The learned class-generic part detectors have the ability to capture the objects across different categories. Experimental results show that the performance of our approach can outperform some recent competing methods.
ER -