University of Twente Student Theses
Analyzing Video Quality Assessment Methods on Computer Graphics Animation Videos
Darici, F.E. (2023) Analyzing Video Quality Assessment Methods on Computer Graphics Animation Videos.
|
PDF
18MB |
| Abstract: | The continuous development of video streaming technologies has created a great demand for accurate assessment of video quality to increase users’ quality of experience (QoE). There are many different categories of videos for user preference such as documentaries, animations, games, and virtual reality (VR) videos. Regardless of the category, every video must go through a video quality assessment to reach the preferred quality by the human visual system (HVS). Thus, the relationship between Video Quality Assessment (VQA) scores and subjective judgments on the quality of videos is open for evaluation in order to improve the overall QoE for users. In this study, we explore the performance of the content-oriented VQA methods on computer graphics (CG) animation videos, since recent VQA studies mainly focus on in-the-wild user-generated-content (UGC) videos. Firstly, we use mean opinion scores (MOS) from human subject opinions on the quality of videos as the baseline to understand the subjective human judgments on the quality of CG animation videos. Secondly, we use videos from the CG Animation Subjective Dataset which are animation and gaming videos exclusively. Thirdly, we compare the state-of-the-art VQA scores on CG videos to mean opinion scores on CG videos to obtain the VQA methods’ performance on CG videos by calculating Spearman’s Rank Correlation Coefficient (SRCC) of the methods’ scores. The results of this study indicate the performance of recent VQA methods on CG animation videos compared to mean opinion scores and propose potential future research directions, such as exploring different VQA methodologies. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Awards: | Best Paper Award |
| Link to this item: | https://purl.utwente.nl/essays/96022 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page