Combining Top-down and Bottom-up Visual Saliency for Firearms Localization

Marco La Cascia; Edoardo Ardizzone; Roberto Gallea; Giuseppe Mazzola

Combining Top-down and Bottom-up Visual Saliency for Firearms Localization

Marco La Cascia, Edoardo Ardizzone, Roberto Gallea, Giuseppe Mazzola

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Object detection is one of the most challenging issues for computer vision researchers. The analysis of thehuman visual attention mechanisms can help automatic inspection systems, in order to discard useless informationand improving performances and efficiency. In this paper we proposed our attention based method toestimate firearms position in images of people holding firearms. Both top-down and bottom-up mechanismsare involved in our system. The bottom-up analysis is based on a state-of-the-art approach. The top-down analysis is based on the construction of a probabilistic model of the firearms position with respect to the people’s face position. This model has been created by analyzing information from of a public available database of movie frames representing actors holding firearms.

Original language	English
Title of host publication	SIGMAP 2014 - International Conference on Signal Processing and Multimedia Applications
Pages	25-32
Number of pages	8
Publication status	Published - 2014

All Science Journal Classification (ASJC) codes

Signal Processing
Computer Graphics and Computer-Aided Design

Access to Document

OA Link

Fingerprint Dive into the research topics of 'Combining Top-down and Bottom-up Visual Saliency for Firearms Localization'. Together they form a unique fingerprint.

Cite this

@inproceedings{ddc5cc6cd52d4a3bbfeeb0c42075f98d,

title = "Combining Top-down and Bottom-up Visual Saliency for Firearms Localization",

abstract = "Object detection is one of the most challenging issues for computer vision researchers. The analysis of thehuman visual attention mechanisms can help automatic inspection systems, in order to discard useless informationand improving performances and efficiency. In this paper we proposed our attention based method toestimate firearms position in images of people holding firearms. Both top-down and bottom-up mechanismsare involved in our system. The bottom-up analysis is based on a state-of-the-art approach. The top-down analysis is based on the construction of a probabilistic model of the firearms position with respect to the people{\textquoteright}s face position. This model has been created by analyzing information from of a public available database of movie frames representing actors holding firearms.",

author = "{La Cascia}, Marco and Edoardo Ardizzone and Roberto Gallea and Giuseppe Mazzola",

year = "2014",

language = "English",

isbn = "978-989-758-046-8",

pages = "25--32",

booktitle = "SIGMAP 2014 - International Conference on Signal Processing and Multimedia Applications",

}

TY - GEN

T1 - Combining Top-down and Bottom-up Visual Saliency for Firearms Localization

AU - La Cascia, Marco

AU - Ardizzone, Edoardo

AU - Gallea, Roberto

AU - Mazzola, Giuseppe

PY - 2014

Y1 - 2014

N2 - Object detection is one of the most challenging issues for computer vision researchers. The analysis of thehuman visual attention mechanisms can help automatic inspection systems, in order to discard useless informationand improving performances and efficiency. In this paper we proposed our attention based method toestimate firearms position in images of people holding firearms. Both top-down and bottom-up mechanismsare involved in our system. The bottom-up analysis is based on a state-of-the-art approach. The top-down analysis is based on the construction of a probabilistic model of the firearms position with respect to the people’s face position. This model has been created by analyzing information from of a public available database of movie frames representing actors holding firearms.

AB - Object detection is one of the most challenging issues for computer vision researchers. The analysis of thehuman visual attention mechanisms can help automatic inspection systems, in order to discard useless informationand improving performances and efficiency. In this paper we proposed our attention based method toestimate firearms position in images of people holding firearms. Both top-down and bottom-up mechanismsare involved in our system. The bottom-up analysis is based on a state-of-the-art approach. The top-down analysis is based on the construction of a probabilistic model of the firearms position with respect to the people’s face position. This model has been created by analyzing information from of a public available database of movie frames representing actors holding firearms.

UR - http://hdl.handle.net/10447/98632

M3 - Conference contribution

SN - 978-989-758-046-8

SP - 25

EP - 32

BT - SIGMAP 2014 - International Conference on Signal Processing and Multimedia Applications

ER -