No-Reference Image Quality Assessment on LIVE-IQA (test)
0.968SROCCDB-CNN
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DB-CNNModel Type=Supervised pretraining and supervised fine-tuning2021.10 | 0.968 | 0.971 | |
| PQRModel Type=Supervised pretraining and supervised fine-tuning2021.10 | 0.965 | 0.971 | |
| HyperIQAModel Type=Supervised pretraining and supervised fine-tuning2021.10 | 0.962 | 0.966 | |
| BIECONModel Type=Supervised pretraining and supervised fine-tuning2021.10 | 0.961 | 0.962 | |
| CONTRIQUEModel Type=Unsupervised pretraining and Linear Regression2021.10 | 0.96 | 0.961 | |
| CORNIAModel Type=Codebook-based Features2021.10 | 0.947 | 0.95 | |
| HOSAModel Type=Codebook-based Features2021.10 | 0.946 | 0.95 | |
| BRISQUEModel Type=Traditional/Handcrafted Features2021.10 | 0.939 | 0.935 | |
| Resnet-50Model Type=Supervised pretraining and Linear Regression2021.10 | 0.925 | 0.931 | |
| NIQEModel Type=Traditional/Handcrafted Features2021.10 | 0.907 | 0.901 |