No Reference Evaluation in Super-Resolution for CCTV Footage

Adit Chawdhary, Seema Kumari, Atnav Bhavsar, Rajesh Verma

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

Abstract

In this paper, we consider the super-resolution (SR) problem in a real scenario involving CCTV videos. As high-resolution (HR) ground-truth is usually not available for such an application, we suggest using no reference image quality assessment metric (e.g. BRISQUE, in this work [1], [20]), to evaluate two example single-image SR approaches suitable for such a task. Notwithstanding the unavailability of ground-truth, we also demonstrate that such a no-reference quality metric, is also more reliable in evaluating SR methods than popular metrics such as PSNR, SSIM even if a separate standard data with HR-LR pair is used to evaluate SR approaches. Finally, some real examples for CCTV images are shown for comparison of SR approaches using BRISQUE.

Original languageEnglish
Title of host publication2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages107-112
Number of pages6
ISBN (Electronic)9781538616765
DOIs
Publication statusPublished - 23-05-2019
Externally publishedYes
Event13th International Conference on Industrial and Information Systems, ICIIS 2018 - Rupnagar, Punjab, India
Duration: 01-12-201802-12-2018

Publication series

Name2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings

Conference

Conference13th International Conference on Industrial and Information Systems, ICIIS 2018
CountryIndia
CityRupnagar, Punjab
Period01-12-1802-12-18

Fingerprint

Closed circuit television systems
Image quality
Evaluation

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Information Systems
  • Information Systems and Management
  • Industrial and Manufacturing Engineering
  • Artificial Intelligence

Cite this

Chawdhary, A., Kumari, S., Bhavsar, A., & Verma, R. (2019). No Reference Evaluation in Super-Resolution for CCTV Footage. In 2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings (pp. 107-112). [8721319] (2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICIINFS.2018.8721319
Chawdhary, Adit ; Kumari, Seema ; Bhavsar, Atnav ; Verma, Rajesh. / No Reference Evaluation in Super-Resolution for CCTV Footage. 2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 107-112 (2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings).
@inproceedings{147f22450035462f9892ec40e669ec15,
title = "No Reference Evaluation in Super-Resolution for CCTV Footage",
abstract = "In this paper, we consider the super-resolution (SR) problem in a real scenario involving CCTV videos. As high-resolution (HR) ground-truth is usually not available for such an application, we suggest using no reference image quality assessment metric (e.g. BRISQUE, in this work [1], [20]), to evaluate two example single-image SR approaches suitable for such a task. Notwithstanding the unavailability of ground-truth, we also demonstrate that such a no-reference quality metric, is also more reliable in evaluating SR methods than popular metrics such as PSNR, SSIM even if a separate standard data with HR-LR pair is used to evaluate SR approaches. Finally, some real examples for CCTV images are shown for comparison of SR approaches using BRISQUE.",
author = "Adit Chawdhary and Seema Kumari and Atnav Bhavsar and Rajesh Verma",
year = "2019",
month = "5",
day = "23",
doi = "10.1109/ICIINFS.2018.8721319",
language = "English",
series = "2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "107--112",
booktitle = "2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings",
address = "United States",

}

Chawdhary, A, Kumari, S, Bhavsar, A & Verma, R 2019, No Reference Evaluation in Super-Resolution for CCTV Footage. in 2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings., 8721319, 2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 107-112, 13th International Conference on Industrial and Information Systems, ICIIS 2018, Rupnagar, Punjab, India, 01-12-18. https://doi.org/10.1109/ICIINFS.2018.8721319

No Reference Evaluation in Super-Resolution for CCTV Footage. / Chawdhary, Adit; Kumari, Seema; Bhavsar, Atnav; Verma, Rajesh.

2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 107-112 8721319 (2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings).

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

TY - GEN

T1 - No Reference Evaluation in Super-Resolution for CCTV Footage

AU - Chawdhary, Adit

AU - Kumari, Seema

AU - Bhavsar, Atnav

AU - Verma, Rajesh

PY - 2019/5/23

Y1 - 2019/5/23

N2 - In this paper, we consider the super-resolution (SR) problem in a real scenario involving CCTV videos. As high-resolution (HR) ground-truth is usually not available for such an application, we suggest using no reference image quality assessment metric (e.g. BRISQUE, in this work [1], [20]), to evaluate two example single-image SR approaches suitable for such a task. Notwithstanding the unavailability of ground-truth, we also demonstrate that such a no-reference quality metric, is also more reliable in evaluating SR methods than popular metrics such as PSNR, SSIM even if a separate standard data with HR-LR pair is used to evaluate SR approaches. Finally, some real examples for CCTV images are shown for comparison of SR approaches using BRISQUE.

AB - In this paper, we consider the super-resolution (SR) problem in a real scenario involving CCTV videos. As high-resolution (HR) ground-truth is usually not available for such an application, we suggest using no reference image quality assessment metric (e.g. BRISQUE, in this work [1], [20]), to evaluate two example single-image SR approaches suitable for such a task. Notwithstanding the unavailability of ground-truth, we also demonstrate that such a no-reference quality metric, is also more reliable in evaluating SR methods than popular metrics such as PSNR, SSIM even if a separate standard data with HR-LR pair is used to evaluate SR approaches. Finally, some real examples for CCTV images are shown for comparison of SR approaches using BRISQUE.

UR - http://www.scopus.com/inward/record.url?scp=85067211658&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067211658&partnerID=8YFLogxK

U2 - 10.1109/ICIINFS.2018.8721319

DO - 10.1109/ICIINFS.2018.8721319

M3 - Conference contribution

T3 - 2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings

SP - 107

EP - 112

BT - 2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Chawdhary A, Kumari S, Bhavsar A, Verma R. No Reference Evaluation in Super-Resolution for CCTV Footage. In 2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 107-112. 8721319. (2018 13th International Conference on Industrial and Information Systems, ICIIS 2018 - Proceedings). https://doi.org/10.1109/ICIINFS.2018.8721319