Compression Based on a Joint Task-Specific Information Metric

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Compression is a key component in many imaging systems in order to accommodate limited resources such as power and bandwidth. Image compression is often done independent of the specific tasks that the systems are designed for, such as target detection, classification, diagnosis, etc. Standard compression techniques are designed based on quality metrics such as mean-squared error (MSE) or peak signal to noise ratio (PSNR). Recently, a metric based on task-specific information (TSI) was proposed and successfully incorporated into JPEG2000 encoding. It has been shown that the proposed TSI metric can optimize the task performance. In this work, a joint metric is proposed to provide a seamless transition between the conventional quality metric MSE and the recently proposed TSI. We demonstrate the effectiveness and flexibility of the proposed joint TSI metric for target detection tasks. Furthermore, it is extended to video tracking applications to demonstrate the robustness of the proposed metric. Experimental results show that although the metric is not directly designed for the applied task, better tracking performance can still be achieved when the joint metric is used, compared to results obtained with the traditional MSE metric.

Original languageEnglish (US)
Title of host publicationData Compression Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages467
Number of pages1
Volume2015-July
ISBN (Print)9781479984305
DOIs
StatePublished - Jul 2 2015
Event2015 Data Compression Conference, DCC 2015 - Snowbird, United States
Duration: Apr 7 2015Apr 9 2015

Other

Other2015 Data Compression Conference, DCC 2015
CountryUnited States
CitySnowbird
Period4/7/154/9/15

Fingerprint

Target tracking
Image compression
Imaging systems
Signal to noise ratio
Bandwidth

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Pu, L., Marcellin, M. W., Bilgin, A., & Ashok, A. (2015). Compression Based on a Joint Task-Specific Information Metric. In Data Compression Conference Proceedings (Vol. 2015-July, pp. 467). [7149330] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DCC.2015.76

Compression Based on a Joint Task-Specific Information Metric. / Pu, Lingling; Marcellin, Michael W; Bilgin, Ali; Ashok, Amit.

Data Compression Conference Proceedings. Vol. 2015-July Institute of Electrical and Electronics Engineers Inc., 2015. p. 467 7149330.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Pu, L, Marcellin, MW, Bilgin, A & Ashok, A 2015, Compression Based on a Joint Task-Specific Information Metric. in Data Compression Conference Proceedings. vol. 2015-July, 7149330, Institute of Electrical and Electronics Engineers Inc., pp. 467, 2015 Data Compression Conference, DCC 2015, Snowbird, United States, 4/7/15. https://doi.org/10.1109/DCC.2015.76
Pu L, Marcellin MW, Bilgin A, Ashok A. Compression Based on a Joint Task-Specific Information Metric. In Data Compression Conference Proceedings. Vol. 2015-July. Institute of Electrical and Electronics Engineers Inc. 2015. p. 467. 7149330 https://doi.org/10.1109/DCC.2015.76
Pu, Lingling ; Marcellin, Michael W ; Bilgin, Ali ; Ashok, Amit. / Compression Based on a Joint Task-Specific Information Metric. Data Compression Conference Proceedings. Vol. 2015-July Institute of Electrical and Electronics Engineers Inc., 2015. pp. 467
@inproceedings{9830838679da4bfaae49e6fdb1e674b6,
title = "Compression Based on a Joint Task-Specific Information Metric",
abstract = "Compression is a key component in many imaging systems in order to accommodate limited resources such as power and bandwidth. Image compression is often done independent of the specific tasks that the systems are designed for, such as target detection, classification, diagnosis, etc. Standard compression techniques are designed based on quality metrics such as mean-squared error (MSE) or peak signal to noise ratio (PSNR). Recently, a metric based on task-specific information (TSI) was proposed and successfully incorporated into JPEG2000 encoding. It has been shown that the proposed TSI metric can optimize the task performance. In this work, a joint metric is proposed to provide a seamless transition between the conventional quality metric MSE and the recently proposed TSI. We demonstrate the effectiveness and flexibility of the proposed joint TSI metric for target detection tasks. Furthermore, it is extended to video tracking applications to demonstrate the robustness of the proposed metric. Experimental results show that although the metric is not directly designed for the applied task, better tracking performance can still be achieved when the joint metric is used, compared to results obtained with the traditional MSE metric.",
author = "Lingling Pu and Marcellin, {Michael W} and Ali Bilgin and Amit Ashok",
year = "2015",
month = "7",
day = "2",
doi = "10.1109/DCC.2015.76",
language = "English (US)",
isbn = "9781479984305",
volume = "2015-July",
pages = "467",
booktitle = "Data Compression Conference Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Compression Based on a Joint Task-Specific Information Metric

AU - Pu, Lingling

AU - Marcellin, Michael W

AU - Bilgin, Ali

AU - Ashok, Amit

PY - 2015/7/2

Y1 - 2015/7/2

N2 - Compression is a key component in many imaging systems in order to accommodate limited resources such as power and bandwidth. Image compression is often done independent of the specific tasks that the systems are designed for, such as target detection, classification, diagnosis, etc. Standard compression techniques are designed based on quality metrics such as mean-squared error (MSE) or peak signal to noise ratio (PSNR). Recently, a metric based on task-specific information (TSI) was proposed and successfully incorporated into JPEG2000 encoding. It has been shown that the proposed TSI metric can optimize the task performance. In this work, a joint metric is proposed to provide a seamless transition between the conventional quality metric MSE and the recently proposed TSI. We demonstrate the effectiveness and flexibility of the proposed joint TSI metric for target detection tasks. Furthermore, it is extended to video tracking applications to demonstrate the robustness of the proposed metric. Experimental results show that although the metric is not directly designed for the applied task, better tracking performance can still be achieved when the joint metric is used, compared to results obtained with the traditional MSE metric.

AB - Compression is a key component in many imaging systems in order to accommodate limited resources such as power and bandwidth. Image compression is often done independent of the specific tasks that the systems are designed for, such as target detection, classification, diagnosis, etc. Standard compression techniques are designed based on quality metrics such as mean-squared error (MSE) or peak signal to noise ratio (PSNR). Recently, a metric based on task-specific information (TSI) was proposed and successfully incorporated into JPEG2000 encoding. It has been shown that the proposed TSI metric can optimize the task performance. In this work, a joint metric is proposed to provide a seamless transition between the conventional quality metric MSE and the recently proposed TSI. We demonstrate the effectiveness and flexibility of the proposed joint TSI metric for target detection tasks. Furthermore, it is extended to video tracking applications to demonstrate the robustness of the proposed metric. Experimental results show that although the metric is not directly designed for the applied task, better tracking performance can still be achieved when the joint metric is used, compared to results obtained with the traditional MSE metric.

UR - http://www.scopus.com/inward/record.url?scp=84938928359&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84938928359&partnerID=8YFLogxK

U2 - 10.1109/DCC.2015.76

DO - 10.1109/DCC.2015.76

M3 - Conference contribution

SN - 9781479984305

VL - 2015-July

SP - 467

BT - Data Compression Conference Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -