Back to search

FRINATEK-Fri prosj.st. mat.,naturv.,tek

Quality and Content: understanding the influence of content on subjective and objective image quality assessment

Alternative title: Kvalitet og innhold: forstå påvirkningen av innhold på subjektiv og objektiv vurdering av bildekvalitet

Awarded: NOK 11.9 mill.

Digital imaging and image processing technologies have changed how we capture, store, transmit, receive, visualize and use images. Advances in technology has encouraged the use of larger volumes of images. A crucial component of any system that processes images is the assessment of the quality. Despite the research efforts on image quality assessment there are still several aspects that are still not fully understood. One property that could affect the overall quality of an image is its content. Although some studies have been performed on this aspect on limited datasets, compared to other properties that could influence the quality of an image, it has been given less attention. Hence, there is a need for better understanding of how content influences subjective quality. Till today, information about content has rarely been included in image quality metrics since how content influences quality is not well understood. In this work, we will address the following question; "How does content influence image quality assessment?". We believe that by investigating this question we can carry out more accurate subjective and objective quality assessment. The goal of the project is to advance the understanding of image quality.

Digital imaging and image processing technologies have changed how we capture, store, transmit, receive, visualize and use images. Advances in technology has encouraged the use of larger volumes of images. A crucial component of any system that processes images is the assessment of the quality. Despite the research efforts on image quality assessment there are still several aspects that are still not fully understood. One property that could affect the overall quality of an image is its content. Although some studies have been performed on this aspect on limited datasets, compared to other properties that could influence the quality of an image, it has been given less attention. Hence, there is a need for better understanding of how content influences subjective quality. Till today, information about content has rarely been included in image quality metrics since how content influences quality is not well understood. In this work, we will address the following question; "How does content influence image quality assessment?". We believe that by investigating this question we can carry out more accurate subjective and objective quality assessment. The goal of the project is to advance the understanding of image quality. This will be done by a focus on: 1) Collecting ground truth data. 2) Investigating how content influences subjective image quality assessment, and 3) Developing more precise and better performing image quality metrics based on the improved understanding on how content influences image quality. In addition to recruiting a new PhD student and one researcher holding a PhD, we include a significant international and inter-disciplinary collaboration with world-leading scholars in perception, image quality assessment and color imaging.

Publications from Cristin

No publications found

No publications found

No publications found

No publications found

Funding scheme:

FRINATEK-Fri prosj.st. mat.,naturv.,tek