List your company
ResearchFormerly WizWand

No-Reference Image Quality Assessment on KonIQ (test)

0.906SROCCHyperIQA
0.5160.617250.71850.81975Aug 16, 2021Aug 27, 2021Sep 8, 2021Sep 20, 2021Oct 1, 2021Oct 13, 2021Oct 25, 2021
Updated 8mo ago

Evaluation results

MethodLinks
HyperIQAModel Type=Supervised pretraining and supervised fine-tuning2021.100.9060.917
CONTRIQUEModel Type=Unsupervised pretraining and Linear Regression2021.100.8940.906
ResNet-50Model Type=Supervised pretraining and Linear Regression2021.100.8880.904
PQRModel Type=Supervised pretraining and supervised fine-tuning2021.100.880.884
DB-CNNModel Type=Supervised pretraining and supervised fine-tuning2021.100.8750.884
PaQ-2-PiQModel Type=Supervised pretraining and supervised fine-tuning2021.100.870.88
HOSAModel Type=Codebook-based Features2021.100.8050.813
CORNIAModel Type=Codebook-based Features2021.100.780.795
HyperIQATrain on dataset=CLIVE, Cross-dataset evaluation protocol=true2021.080.772—
HyperIQATrain on dataset=LIVEFB, Cross-dataset evaluation protocol=true2021.080.758—
P2P-BMTrain on dataset=LIVEFB, Cross-dataset evaluation protocol=true2021.080.755—
DBCNNTrain on dataset=CLIVE, Cross-dataset evaluation protocol=true2021.080.754—
P2P-BMTrain on dataset=CLIVE, Cross-dataset evaluation protocol=true2021.080.74—
TReSTrain on dataset=CLIVE, Cross-dataset evaluation protocol=true2021.080.733—
DBCNNTrain on dataset=LIVEFB, Cross-dataset evaluation protocol=true2021.080.716—
TReSTrain on dataset=LIVEFB, Cross-dataset evaluation protocol=true2021.080.713—
WaDIQaMTrain on dataset=CLIVE, Cross-dataset evaluation protocol=true2021.080.711—
WaDIQaMTrain on dataset=LIVEFB, Cross-dataset evaluation protocol=true2021.080.708—
BRISQUEModel Type=Traditional/Handcrafted Features2021.100.6650.681
NIQEModel Type=Traditional/Handcrafted Features2021.100.5310.538