Evaluation Infrastructure for the Measurement of Content-based Video Quality and Video Analytics Performance

Published

July 31, 2017

Author(s)

Haiying Guan, Daniel F. Zhou, Jonathan G. Fiscus, John S. Garofolo, James M. Horan

Abstract

In this report we describe a novel framework for evaluation of video analytics which uses video quality metric prediction as a measurement of the system performance. This system is used to improve overall video analytics system performance. The primary concept of this approach is to create an evaluation infrastructure that can promote the study of video quality metrics, and at the same time promote the research of video analytics. The achievements in both domains will have significant impacts on many application areas, such as public safety, homeland security, video surveillance applications, geographic applications and entertainment video.

Citation

NIST Interagency/Internal Report (NISTIR) - 8187

Report Number

8187

NIST Pub Series

NIST Interagency/Internal Report (NISTIR)

Pub Type

NIST Pubs

Download Paper

DOI Link

Keywords

Video Quality Metric (VQM), Video Analytics (VA), Multimedia Event Detection (MED), TREC Video Retrieval Evaluation (TRECVID), Point and Shoot Face Recognition Challenge (PaSC), Face Recognition, Detection Analysis Pipeline Resources (DAPR), Scorer-centric Evaluation, Analysis- Centric Evaluation, National Imagery Interpretability Rating Scale (NIIRS), Video-National Imagery Interpretability Rating Scale (VNIIRS), Mean opinion score (MOS), Difference Mean opinion score (DMOS).

Citation

Guan, H. , Zhou, D. , Fiscus, J. , Garofolo, J. and Horan, J. (2017), Evaluation Infrastructure for the Measurement of Content-based Video Quality and Video Analytics Performance, NIST Interagency/Internal Report (NISTIR), National Institute of Standards and Technology, Gaithersburg, MD, [online], https://doi.org/10.6028/NIST.IR.8187 (Accessed November 21, 2024)