Pedestrian detection with unsupervised multi-stage feature learning

Pierre Sermanet; Koray Kavukcuoglu; Soumith Chintala; Yann Lecun

doi:10.1109/CVPR.2013.465

Pedestrian detection with unsupervised multi-stage feature learning

Pierre Sermanet, Koray Kavukcuoglu, Soumith Chintala, Yann Lecun

Computer Science

Research output: Contribution to journal › Conference article › peer-review

Abstract

Pedestrian detection is a problem of considerable practical interest. Adding to the list of successful applications of deep learning methods to vision, we report state-of-the-art and competitive results on all major pedestrian datasets with a convolutional network model. The model uses a few new twists, such as multi-stage features, connections that skip layers to integrate global shape information with local distinctive motif information, and an unsupervised method based on convolutional sparse coding to pre-train the filters at each stage.

Original language	English (US)
Article number	6619309
Pages (from-to)	3626-3633
Number of pages	8
Journal	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
DOIs	https://doi.org/10.1109/CVPR.2013.465
State	Published - 2013
Event	26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2013 - Portland, OR, United States Duration: Jun 23 2013 → Jun 28 2013

Keywords

computer vision
convolutional
deep learning
detection
pedestrian
unsupervised

ASJC Scopus subject areas

Software
Computer Vision and Pattern Recognition

Access to Document

10.1109/CVPR.2013.465

Cite this

@article{82294fc46f3a4f75aa85a3f344b0358a,

title = "Pedestrian detection with unsupervised multi-stage feature learning",

abstract = "Pedestrian detection is a problem of considerable practical interest. Adding to the list of successful applications of deep learning methods to vision, we report state-of-the-art and competitive results on all major pedestrian datasets with a convolutional network model. The model uses a few new twists, such as multi-stage features, connections that skip layers to integrate global shape information with local distinctive motif information, and an unsupervised method based on convolutional sparse coding to pre-train the filters at each stage.",

keywords = "computer vision, convolutional, deep learning, detection, pedestrian, unsupervised",

author = "Pierre Sermanet and Koray Kavukcuoglu and Soumith Chintala and Yann Lecun",

year = "2013",

doi = "10.1109/CVPR.2013.465",

language = "English (US)",

pages = "3626--3633",

journal = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

issn = "1063-6919",

publisher = "IEEE Computer Society",

note = "26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2013 ; Conference date: 23-06-2013 Through 28-06-2013",

}

TY - JOUR

T1 - Pedestrian detection with unsupervised multi-stage feature learning

AU - Sermanet, Pierre

AU - Kavukcuoglu, Koray

AU - Chintala, Soumith

AU - Lecun, Yann

PY - 2013

Y1 - 2013

N2 - Pedestrian detection is a problem of considerable practical interest. Adding to the list of successful applications of deep learning methods to vision, we report state-of-the-art and competitive results on all major pedestrian datasets with a convolutional network model. The model uses a few new twists, such as multi-stage features, connections that skip layers to integrate global shape information with local distinctive motif information, and an unsupervised method based on convolutional sparse coding to pre-train the filters at each stage.

AB - Pedestrian detection is a problem of considerable practical interest. Adding to the list of successful applications of deep learning methods to vision, we report state-of-the-art and competitive results on all major pedestrian datasets with a convolutional network model. The model uses a few new twists, such as multi-stage features, connections that skip layers to integrate global shape information with local distinctive motif information, and an unsupervised method based on convolutional sparse coding to pre-train the filters at each stage.

KW - computer vision

KW - convolutional

KW - deep learning

KW - detection

KW - pedestrian

KW - unsupervised

UR - http://www.scopus.com/inward/record.url?scp=84887328988&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887328988&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2013.465

DO - 10.1109/CVPR.2013.465

M3 - Conference article

AN - SCOPUS:84887328988

SN - 1063-6919

SP - 3626

EP - 3633

JO - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

JF - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

M1 - 6619309

T2 - 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2013

Y2 - 23 June 2013 through 28 June 2013

ER -

Pedestrian detection with unsupervised multi-stage feature learning

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this