Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A search space strategy for pedestrian detection and localization in world coordinates

Nilsson, Mikael LU ; Ahrnbom, Martin LU orcid ; Ardo, Håkan LU and Laureshyn, Aliaksei LU orcid (2018) 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2018 5. p.17-24
Abstract

The focus of this work is detecting pedestrians, captured in a surveillance setting, and locating them in world coordinates. Commonly adopted search strategies operate in the image plane to address the object detection problem with machine learning, for example using scale-space pyramid with the sliding windows methodology or object proposals. In contrast, here a new search space is presented, which exploits camera calibration information and geometric priors. The proposed search strategy will facilitate detectors to directly estimate pedestrian presence in world coordinates of interest. Results are demonstrated on real world outdoor collected data along a path in dim light conditions, with the goal to locate pedestrians in world... (More)

The focus of this work is detecting pedestrians, captured in a surveillance setting, and locating them in world coordinates. Commonly adopted search strategies operate in the image plane to address the object detection problem with machine learning, for example using scale-space pyramid with the sliding windows methodology or object proposals. In contrast, here a new search space is presented, which exploits camera calibration information and geometric priors. The proposed search strategy will facilitate detectors to directly estimate pedestrian presence in world coordinates of interest. Results are demonstrated on real world outdoor collected data along a path in dim light conditions, with the goal to locate pedestrians in world coordinates. The proposed search strategy indicate a mean error under 20 cm, while image plane search methods, with additional processing adopted for localization, yielded around or above 30 cm in mean localization error. This while only observing 3-4% of patches required by the image plane searches at the same task.

(Less)
Please use this url to cite or link to this publication:
author
; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Camera calibration, Detection, Machine learning, Pedestrian, World coordinates
host publication
VISAPP
volume
5
pages
8 pages
publisher
SciTePress
conference name
13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2018
conference location
Funchal, Madeira, Portugal
conference dates
2018-01-27 - 2018-01-29
external identifiers
  • scopus:85047811507
ISBN
9789897582905
language
English
LU publication?
yes
id
822d67f7-3cde-45d1-8809-eb4b89f02597
date added to LUP
2018-06-15 13:13:59
date last changed
2022-05-03 03:44:55
@inproceedings{822d67f7-3cde-45d1-8809-eb4b89f02597,
  abstract     = {{<p>The focus of this work is detecting pedestrians, captured in a surveillance setting, and locating them in world coordinates. Commonly adopted search strategies operate in the image plane to address the object detection problem with machine learning, for example using scale-space pyramid with the sliding windows methodology or object proposals. In contrast, here a new search space is presented, which exploits camera calibration information and geometric priors. The proposed search strategy will facilitate detectors to directly estimate pedestrian presence in world coordinates of interest. Results are demonstrated on real world outdoor collected data along a path in dim light conditions, with the goal to locate pedestrians in world coordinates. The proposed search strategy indicate a mean error under 20 cm, while image plane search methods, with additional processing adopted for localization, yielded around or above 30 cm in mean localization error. This while only observing 3-4% of patches required by the image plane searches at the same task.</p>}},
  author       = {{Nilsson, Mikael and Ahrnbom, Martin and Ardo, Håkan and Laureshyn, Aliaksei}},
  booktitle    = {{VISAPP}},
  isbn         = {{9789897582905}},
  keywords     = {{Camera calibration; Detection; Machine learning; Pedestrian; World coordinates}},
  language     = {{eng}},
  month        = {{01}},
  pages        = {{17--24}},
  publisher    = {{SciTePress}},
  title        = {{A search space strategy for pedestrian detection and localization in world coordinates}},
  volume       = {{5}},
  year         = {{2018}},
}