On the effectiveness of natural videos in masking dynamic DCT noise

Author

Evert, Jeremy ; Chandler, David L.

Author_Institution

Lab. of Comput. Perception & Image Quality (CPIQ), Oklahoma State Univ., Stillwater, OK, USA

fYear

2013

fDate

3-6 Nov. 2013

Firstpage

1339

Lastpage

1345

Abstract

A fundamental understanding of what makes distortions harder to detect in videos can benefit efforts in video processing. We present the results of a study which extends the work of Watson et al. [J. Electronic Imaging, 2001], to investigate masking of dynamic DCT distortions by eight natural videos. Distortions ranged from 0-60 Hz in temporal frequency and from 2-22.6 c/deg in spatial frequency. Our results revealed that detection thresholds are jointly influenced by the spatiotemporal characteristics of the distortion and by the spatiotemporal characteristics of the video. We discuss the implications of these results for video coding and quality assessment.

Keywords

discrete cosine transforms; distortion; natural scenes; spatiotemporal phenomena; video signal processing; dynamic DCT distortions; dynamic DCT noise; frequency 0 Hz to 60 Hz; masking; natural videos; quality assessment; spatial frequency; spatiotemporal characteristics; temporal frequency; video coding; video processing; Discrete cosine transforms; Face; Frequency modulation; Noise; Observers; Standards; Videos;

fLanguage

English

Publisher

ieee

Conference_Titel

Signals, Systems and Computers, 2013 Asilomar Conference on

Conference_Location

Pacific Grove, CA

Print_ISBN

978-1-4799-2388-5

Type

conf

DOI

10.1109/ACSSC.2013.6810512

Filename

6810512