Picture quality prediction on CLIVE (val)
0.784SRCCFeedback Model
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Feedback ModelBackbone=ResNet-18, trained_on=PaQ-2-PiQ, fine-tuning=false2019.12 | 0.784 | 0.754 | |
| RoIPool ModelBackbone=ResNet-18, trained_on=PaQ-2-PiQ, fine-tuning=false2019.12 | 0.762 | 0.775 | |
| Baseline ModelBackbone=ResNet-18, trained_on=PaQ-2-PiQ, fine-tuning=false2019.12 | 0.74 | 0.725 | |
| NIMA(MobileNet v2)Backbone=MobileNet v2, trained_on=PaQ-2-PiQ, fine-tuning=false2019.12 | 0.712 | 0.705 | |
| NIMA(ResNet 18)Backbone=ResNet-18, trained_on=PaQ-2-PiQ, fine-tuning=false2019.12 | 0.707 | 0.645 |