List your company
ResearchFormerly WizWand

No-Reference Image Quality Assessment on LIVE-C

0.903PLCCOurs-S
0.49220.598850.70550.81215Oct 14, 2023Mar 2, 2024Jul 21, 2024Dec 8, 2024Apr 28, 2025Sep 15, 2025Feb 3, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Ours-SBackbone=Swin-Tiny, Joint training (FR & NR tasks)=false, Architecture Type=Transformer-based2023.100.9030.869
MS-SCANet2026.020.9030.895
MAMIQA2026.020.8950.874
Ours-RBackbone=ResNet50, Joint training (FR & NR tasks)=false2023.100.8920.841
HyperIQAArchitecture Type=CNN/Hand-crafted2023.100.8820.859
TRESArchitecture Type=CNN/Hand-crafted2023.100.8770.846
TreS2026.020.8770.846
DBCNNArchitecture Type=CNN/Hand-crafted2023.100.8690.869
DBCNN2026.020.8690.869
TIQAArchitecture Type=Transformer-based2023.100.8610.845
TRIQ2026.020.8610.845
MANIQATraining Dataset=Koniq10k and HiRQA2025.080.8490.832
P2P-BMArchitecture Type=CNN/Hand-crafted2023.100.8420.844
HyperIQA2026.020.8420.844
CLIP-IQA+Training Dataset=Koniq10k and HiRQA2025.080.8320.805
MetaIQAArchitecture Type=CNN/Hand-crafted2023.100.8020.835
MUSIQTraining Dataset=Koniq10k and HiRQA2025.080.7890.83
DBCNNTraining Dataset=Koniq10k and HiRQA2025.080.7730.755
ARNIQATraining Dataset=Koniq10k and HiRQA2025.080.7590.724
HiRQATraining Dataset=Koniq10k and HiRQA2025.080.7440.692
MEONArchitecture Type=CNN/Hand-crafted2023.100.710.697
MEON2026.020.710.697
WaDIQaMArchitecture Type=CNN/Hand-crafted2023.100.6710.682
BRISQUEArchitecture Type=CNN/Hand-crafted2023.100.6290.629
BRISQUE2026.020.6290.629
BIECONArchitecture Type=CNN/Hand-crafted2023.100.6130.613
DIIVINEArchitecture Type=CNN/Hand-crafted2023.100.5910.588
DIIVINE2026.020.5910.588
ILNIQEArchitecture Type=CNN/Hand-crafted2023.100.5080.508