CNN-based Cross-dataset No-reference Image Quality Assessment

Dan Yang, Veli-Tapani Peltoketo, Joni-Kristian Kämäräinen

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

5 Citations (Scopus)
5 Downloads (Pure)


Recent works on no-reference image quality assessment (NR-IQA) have reported good performance for various datasets. However, they suffer from significant performance drops in cross-dataset evaluations which indicates poor generalization power. We propose a Siamese architecture and training procedures for cross-dataset deep NR-IQA that achieves clearly better performance. Moreover, we show that the architecture can be further boosted by i) pre-training with a large aesthetics dataset and ii) adding low-level quality cues, sharpness, tone and colourfulness, as additional features.
Original languageEnglish
Title of host publication2019 International Conference on Computer Vision Workshop, ICCVW 2019
Number of pages9
ISBN (Electronic)978-1-7281-5023-9
ISBN (Print)978-1-7281-5024-6
Publication statusPublished - 2019
Publication typeA4 Article in conference proceedings
EventIEEE International Conference on Computer Vision Workshops -
Duration: 1 Jan 1900 → …

Publication series

NameIEEE International Conference on Computer Vision workshops
ISSN (Print)2473-9936
ISSN (Electronic)2473-9944


ConferenceIEEE International Conference on Computer Vision Workshops
Period1/01/00 → …

Publication forum classification

  • Publication forum level 1


Dive into the research topics of 'CNN-based Cross-dataset No-reference Image Quality Assessment'. Together they form a unique fingerprint.

Cite this