Data-driven street scene layout estimation for distant object detection

Donghao Zhang; Xuming He; Hanxi Li

doi:10.1109/DICTA.2014.7008099

Data-driven street scene layout estimation for distant object detection

Donghao Zhang, Xuming He, Hanxi Li

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

7 Citations (Scopus)

Abstract

We present a street scene layout estimation method based on transferring layout annotation from a (large) image database and its application for distant object detection. Inspired by nonparametric scene labeling approaches, we estimate a scene's geometric layout by matching global image descriptors and retrieving the most similar layout configuration. Our label transfer is done for each sub-region of an image and a tiered scene model is used to integrate all the local label information into a coherent scene layout prediction. Given the geometric layout, we use a super-resolution method to zoom in the distance region and refine the search in object detection. On KITTI dataset, we show that we can reliably generate scene layout and improve the detection of distant cars over the state of the art DPM detector.

Original language	English
Title of host publication	2014 International Conference on Digital Image Computing
Subtitle of host publication	Techniques and Applications, DICTA 2014
Editors	Abdesselam Bouzerdoum, Lei Wang, Philip Ogunbona, Wanqing Li, Son Lam Phung
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	1-7
Number of pages	7
ISBN (Electronic)	9781479954094
DOIs	https://doi.org/10.1109/DICTA.2014.7008099
Publication status	Published - 12 Jan 2015
Externally published	Yes
Event	Digital Image Computing Techniques and Applications 2014 - Wollongong, Australia Duration: 25 Nov 2014 → 27 Nov 2014 Conference number: 16th https://ssl.informatics.uow.edu.au/dicta2014/ https://ieeexplore.ieee.org/xpl/conhome/7005557/proceeding (Proceedings)

Conference

Conference	Digital Image Computing Techniques and Applications 2014
Abbreviated title	DICTA 2014
Country/Territory	Australia
City	Wollongong
Period	25/11/14 → 27/11/14
Internet address	https://ssl.informatics.uow.edu.au/dicta2014/ https://ieeexplore.ieee.org/xpl/conhome/7005557/proceeding (Proceedings)

Access to Document

10.1109/DICTA.2014.7008099

Cite this

Zhang, D., He, X., & Li, H. (2015). Data-driven street scene layout estimation for distant object detection. In A. Bouzerdoum, L. Wang, P. Ogunbona, W. Li, & S. L. Phung (Eds.), 2014 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2014 (pp. 1-7). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/DICTA.2014.7008099

Zhang, Donghao ; He, Xuming ; Li, Hanxi. / Data-driven street scene layout estimation for distant object detection. 2014 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2014. editor / Abdesselam Bouzerdoum ; Lei Wang ; Philip Ogunbona ; Wanqing Li ; Son Lam Phung. IEEE, Institute of Electrical and Electronics Engineers, 2015. pp. 1-7

@inproceedings{fb782d1c6c68439f839277f9e9139161,

title = "Data-driven street scene layout estimation for distant object detection",

abstract = "We present a street scene layout estimation method based on transferring layout annotation from a (large) image database and its application for distant object detection. Inspired by nonparametric scene labeling approaches, we estimate a scene's geometric layout by matching global image descriptors and retrieving the most similar layout configuration. Our label transfer is done for each sub-region of an image and a tiered scene model is used to integrate all the local label information into a coherent scene layout prediction. Given the geometric layout, we use a super-resolution method to zoom in the distance region and refine the search in object detection. On KITTI dataset, we show that we can reliably generate scene layout and improve the detection of distant cars over the state of the art DPM detector.",

author = "Donghao Zhang and Xuming He and Hanxi Li",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; Digital Image Computing Techniques and Applications 2014, DICTA 2014 ; Conference date: 25-11-2014 Through 27-11-2014",

year = "2015",

month = jan,

day = "12",

doi = "10.1109/DICTA.2014.7008099",

language = "English",

pages = "1--7",

editor = "Abdesselam Bouzerdoum and Lei Wang and Philip Ogunbona and Wanqing Li and Phung, {Son Lam}",

booktitle = "2014 International Conference on Digital Image Computing",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

url = "https://ssl.informatics.uow.edu.au/dicta2014/, https://ieeexplore.ieee.org/xpl/conhome/7005557/proceeding",

}

Zhang, D, He, X & Li, H 2015, Data-driven street scene layout estimation for distant object detection. in A Bouzerdoum, L Wang, P Ogunbona, W Li & SL Phung (eds), 2014 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2014. IEEE, Institute of Electrical and Electronics Engineers, pp. 1-7, Digital Image Computing Techniques and Applications 2014, Wollongong, New South Wales, Australia, 25/11/14. https://doi.org/10.1109/DICTA.2014.7008099

Data-driven street scene layout estimation for distant object detection. / Zhang, Donghao; He, Xuming; Li, Hanxi.
2014 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2014. ed. / Abdesselam Bouzerdoum; Lei Wang; Philip Ogunbona; Wanqing Li; Son Lam Phung. IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 1-7.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Data-driven street scene layout estimation for distant object detection

AU - Zhang, Donghao

AU - He, Xuming

AU - Li, Hanxi

N1 - Conference code: 16th

PY - 2015/1/12

Y1 - 2015/1/12

N2 - We present a street scene layout estimation method based on transferring layout annotation from a (large) image database and its application for distant object detection. Inspired by nonparametric scene labeling approaches, we estimate a scene's geometric layout by matching global image descriptors and retrieving the most similar layout configuration. Our label transfer is done for each sub-region of an image and a tiered scene model is used to integrate all the local label information into a coherent scene layout prediction. Given the geometric layout, we use a super-resolution method to zoom in the distance region and refine the search in object detection. On KITTI dataset, we show that we can reliably generate scene layout and improve the detection of distant cars over the state of the art DPM detector.

AB - We present a street scene layout estimation method based on transferring layout annotation from a (large) image database and its application for distant object detection. Inspired by nonparametric scene labeling approaches, we estimate a scene's geometric layout by matching global image descriptors and retrieving the most similar layout configuration. Our label transfer is done for each sub-region of an image and a tiered scene model is used to integrate all the local label information into a coherent scene layout prediction. Given the geometric layout, we use a super-resolution method to zoom in the distance region and refine the search in object detection. On KITTI dataset, we show that we can reliably generate scene layout and improve the detection of distant cars over the state of the art DPM detector.

UR - http://www.scopus.com/inward/record.url?scp=84922575288&partnerID=8YFLogxK

U2 - 10.1109/DICTA.2014.7008099

DO - 10.1109/DICTA.2014.7008099

M3 - Conference Paper

AN - SCOPUS:84922575288

SP - 1

EP - 7

BT - 2014 International Conference on Digital Image Computing

A2 - Bouzerdoum, Abdesselam

A2 - Wang, Lei

A2 - Ogunbona, Philip

A2 - Li, Wanqing

A2 - Phung, Son Lam

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - Digital Image Computing Techniques and Applications 2014

Y2 - 25 November 2014 through 27 November 2014

ER -

Data-driven street scene layout estimation for distant object detection

Abstract

Conference

Access to Document

Other files and links

Cite this