Fast compressed domain motion detection in H.264 video streams for video surveillance applications

Krzysztof Szczerba; Søren Forchhammer; Jesper Støttrup-Andersen; Peder Tanderup Eybye

doi:10.1109/AVSS.2009.78

Fast compressed domain motion detection in H.264 video streams for video surveillance applications

Krzysztof Szczerba, Søren Forchhammer, Jesper Støttrup-Andersen, Peder Tanderup Eybye

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

3137 Downloads (Pure)

Abstract

This paper presents a novel approach to fast motion detection in H.264/MPEG-4 advanced video coding (AVC) compressed video streams for IP video surveillance systems. The goal is to develop algorithms which may be useful in a real-life industrial perspective by facilitating the processing of large numbers of video streams on a single server. The focus of the work is on using the information in coded video streams to reduce the computational complexity and memory requirements, which translates into reduced hardware requirements and costs. The devised algorithm detects and segments activity based on motion vectors embedded in the video stream without requiring a full decoding and reconstruction of video frames. To improve the robustness to noise, a confidence measure based on temporal and spatial clues is introduced to increase the probability of correct detection. The algorithm was tested on indoor surveillance H.264 sequences.

Original language	English
Title of host publication	Proceedings, AVSS
Publisher	IEEE
Publication date	2009
Pages	478-483
ISBN (Print)	978-0-7695-3718-4
DOIs	https://doi.org/10.1109/AVSS.2009.78
Publication status	Published - 2009
Event	6th IEEE Conference on Advanced Video and Signal Based Surveillance - Genova, Italy Duration: 2 Sept 2009 → 4 Sept 2009 Conference number: 6

Conference

Conference	6th IEEE Conference on Advanced Video and Signal Based Surveillance
Number	6
Country/Territory	Italy
City	Genova
Period	02/09/2009 → 04/09/2009

Bibliographical note

Copyright: 2009 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

Access to Document

10.1109/AVSS.2009.78

SzczerbaFinal published version, 218 KB

OpenUrl availability

Full text

Cite this

@inproceedings{ee4ffc01e1a94040a7a6b9678f0cfbb9,

title = "Fast compressed domain motion detection in H.264 video streams for video surveillance applications",

abstract = "This paper presents a novel approach to fast motion detection in H.264/MPEG-4 advanced video coding (AVC) compressed video streams for IP video surveillance systems. The goal is to develop algorithms which may be useful in a real-life industrial perspective by facilitating the processing of large numbers of video streams on a single server. The focus of the work is on using the information in coded video streams to reduce the computational complexity and memory requirements, which translates into reduced hardware requirements and costs. The devised algorithm detects and segments activity based on motion vectors embedded in the video stream without requiring a full decoding and reconstruction of video frames. To improve the robustness to noise, a confidence measure based on temporal and spatial clues is introduced to increase the probability of correct detection. The algorithm was tested on indoor surveillance H.264 sequences.",

author = "Krzysztof Szczerba and S{\o}ren Forchhammer and Jesper St{\o}ttrup-Andersen and Eybye, {Peder Tanderup}",

note = "Copyright: 2009 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE; 6th IEEE Conference on Advanced Video and Signal Based Surveillance, AVSS ; Conference date: 02-09-2009 Through 04-09-2009",

year = "2009",

doi = "10.1109/AVSS.2009.78",

language = "English",

isbn = "978-0-7695-3718-4",

pages = "478--483",

booktitle = "Proceedings, AVSS",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - Fast compressed domain motion detection in H.264 video streams for video surveillance applications

AU - Szczerba, Krzysztof

AU - Forchhammer, Søren

AU - Støttrup-Andersen, Jesper

AU - Eybye, Peder Tanderup

N1 - Conference code: 6

PY - 2009

Y1 - 2009

N2 - This paper presents a novel approach to fast motion detection in H.264/MPEG-4 advanced video coding (AVC) compressed video streams for IP video surveillance systems. The goal is to develop algorithms which may be useful in a real-life industrial perspective by facilitating the processing of large numbers of video streams on a single server. The focus of the work is on using the information in coded video streams to reduce the computational complexity and memory requirements, which translates into reduced hardware requirements and costs. The devised algorithm detects and segments activity based on motion vectors embedded in the video stream without requiring a full decoding and reconstruction of video frames. To improve the robustness to noise, a confidence measure based on temporal and spatial clues is introduced to increase the probability of correct detection. The algorithm was tested on indoor surveillance H.264 sequences.

AB - This paper presents a novel approach to fast motion detection in H.264/MPEG-4 advanced video coding (AVC) compressed video streams for IP video surveillance systems. The goal is to develop algorithms which may be useful in a real-life industrial perspective by facilitating the processing of large numbers of video streams on a single server. The focus of the work is on using the information in coded video streams to reduce the computational complexity and memory requirements, which translates into reduced hardware requirements and costs. The devised algorithm detects and segments activity based on motion vectors embedded in the video stream without requiring a full decoding and reconstruction of video frames. To improve the robustness to noise, a confidence measure based on temporal and spatial clues is introduced to increase the probability of correct detection. The algorithm was tested on indoor surveillance H.264 sequences.

U2 - 10.1109/AVSS.2009.78

DO - 10.1109/AVSS.2009.78

M3 - Article in proceedings

SN - 978-0-7695-3718-4

SP - 478

EP - 483

BT - Proceedings, AVSS

PB - IEEE

T2 - 6th IEEE Conference on Advanced Video and Signal Based Surveillance

Y2 - 2 September 2009 through 4 September 2009

ER -

Fast compressed domain motion detection in H.264 video streams for video surveillance applications

Abstract

Conference

Bibliographical note

Access to Document

OpenUrl availability

Fingerprint

Cite this