Quality Assessment for Natural and Screen Visual Contents
| Autor: | David B. L. Bong, Woei-Tan Loh |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Information retrieval
Image quality Feature (computer vision) Computer science Quality assessment 0202 electrical engineering electronic engineering information engineering 020206 networking & telecommunications 020201 artificial intelligence & image processing 02 engineering and technology Visualization |
| Zdroj: | ICIP |
| DOI: | 10.1109/icip.2019.8803254 |
| Popis: | Quality assessment (QA) of screen visual contents (SVCs) has gained more and more popularity. SVCs are very different from natural visual contents (NVCs) which have been dealt with by most researchers in the literature. QA methods specifically designed for NVCs also can be used to evaluate the quality of SVCs. Yet, their performances are unsatisfactory. This may due to the statistical differences of SVCs and NVCs. In the thesis, SVCs and NVCs QA methods in the literature are being compared and studied for both SVCs and NVCs benchmarked databases. It is found out that methods that incorporate gradient features work well for both SVCs and NVCs. Thus, gradient feature is utilized in designing QA method that works for both SVCs and NVCs simultaneously.QA methods will be designed first for natural and screen content images. Then, the methods will be further enhanced for video QA purposes. This is the first attempt in formulating QA methods that work for SVCs and NVCs simultaneously. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|