Stefanie Wechtitsch
Werner Bailer
Lucas Paletta
2018-09-04
<p>Automatic quality assessment is an essential step in the professional audiovisual media production process. In this paper we propose a novel sharpness metric taking the specific properties of video into account, and having a higher robustness against variations in image content, interlacing artifacts and noise.<br>
Furthermore, a comprehensive user study is presented, where we obtain subjective scores to validate the sharpness metric. We ask 28 viewers for both absolute and relative judgments of sharpness in two separate experiment settings, supported with an eye tracker to obtain the locations used for judgements.<br>
Experimental results show that the objective sharpness metric under test is well correlated with human perception. Both results, the absolute and relative subjective ratings confirm the good correlation of the proposed metric with human perception. The analysis of the eye tracking data highlights differences between experts and consumers.</p>
https://doi.org/10.1109/CBMI.2018.8516509
oai:zenodo.org:1492248
eng
Zenodo
https://zenodo.org/communities/marconi_eu
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution Non Commercial No Derivatives 4.0 International
https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode
CBMI, International Conference on Content-based Multimedia Indexing, La Rochelle, FR, 4-6 September 2018
Refining and Evaluating a Video Sharpness Metric
info:eu-repo/semantics/conferencePaper