CNN-Based Cross-Dataset No-Reference Image Quality Assessment

Dan Yang, Veli-Tapani Peltoketo, Joni-Kristian Kamarainen
2019 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)  
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)
more » ... ng low-level quality cues, sharpness, tone and colourfulness, as additional features. Related work NSS-based NR-IQA methods define the problem as a classification or a regression problem for features that represent natural scene statistics (NSS) or statistics learned from data [45] . NSS-based methods assume that there are statis-
doi:10.1109/iccvw.2019.00485 dblp:conf/iccvw/YangPK19 fatcat:72ruk7bhmjcjxabnlbjtfcfrna