Video quality subjective assessment considering cognitive criteria and user preferences on video content

  • Demóstenes Zegarra Rodríguez USP
  • Júlia Abrahão USP
  • Dante Coaquira Begazo USP
  • Renata Lopes Rosa USP
  • Graça Bressan USP


There are several subjective test methodologies for assessing video quality, being that each of these methods was developed by considering specific degradation factors in video quality. Therefore, for video streaming services that use TCP as transport protocol different evaluation criteria should be considered as compared to current methods, because the impairment are only pauses. In this paper, the concepts of the mechanisms that integrate perceptual and cognitive processes are studied, relating them to a test environment for subjective evaluation of video, which is limited to streaming video service transmitted via HTTP / TCP. For this reason, some recommendations to perform subjective assessment of streaming video service are proposed, taking into account perceptual and cognitive criteria. In this work, subjective tests are conducted, evaluators with different content preferences assess 3 different video content. Results show that the assessors' QoE depends strongly on video content preferences. Thus, assessors who are interested in the video content are more drastic in the ratings. In the case of sport videos, the Mean Opinion Score (MOS) value of assesors with explicit preference in sports reachs only 54% of the MOS obtained by assessors without interest in sports. Furthermore, the roll of the instructor during the test is also evaluated.
Como Citar

Selecione um Formato
RODRÍGUEZ, Demóstenes Zegarra; ABRAHÃO, Júlia; BEGAZO, Dante Coaquira; ROSA, Renata Lopes; BRESSAN, Graça. Video quality subjective assessment considering cognitive criteria and user preferences on video content. In: ANAIS PRINCIPAIS DO SIMPÓSIO BRASILEIRO DE SISTEMAS MULTIMÍDIA E WEB (WEBMEDIA), 18. , 2012, São Paulo. Anais Principais do XVIII Simpósio Brasileiro de Sistemas Multimídia e Web. Porto Alegre: Sociedade Brasileira de Computação, oct. 2012 . p. 269-272.

Artigos mais lidos do(s) mesmo(s) autor(es)