List your company
ResearchFormerly WizWand

Image Quality Assessment on AGIQA

0.842SRCCMG-IQA
0.40520.51860.6320.7454Oct 13, 2025Nov 13, 2025Dec 14, 2025Jan 14, 2026Feb 14, 2026Mar 17, 2026Apr 18, 2026
Updated 6mo ago

Evaluation results

MethodLinks
MG-IQAmulti-dataset training=true, training_data=KADID-10k + SPAQ2026.040.8420.858
Q-ProbeCategory=MLLMs-based2026.010.8370.813
VQ-R1multi-dataset training=true, training_data=KADID-10k + SPAQ2026.040.8130.831
VisualQuality-R1Category=MLLMs-based2026.010.8050.843
Q-Insightmulti-dataset training=true, training_data=KADID-10k + SPAQ2026.040.7720.79
Q-DeepSight†Category=MLLM (w/ reasoning), Training Dataset=KonIQ + DQ-7K2026.040.7680.822
VisualQuality-R1Category=MLLM-based, Training Dataset=KonIQ2025.100.7670.822
Q-InsightCategory=MLLM (w/ reasoning), Training Dataset=KonIQ2026.040.7660.816
Zoom-IQACategory=MLLM (w/ reasoning), Training Dataset=KonIQ2026.040.7650.816
Q-InsightCategory=MLLM-based, Training Dataset=KonIQ2025.100.7640.811
Q-Insight†Category=MLLM (w/ reasoning), Training Dataset=KonIQ + DQ-7K2026.040.7640.811
VisualQuality-R1Category=MLLM (w/ reasoning), Training Dataset=KonIQ2026.040.760.817
Q-DeepSightCategory=MLLM (w/ reasoning), Training Dataset=KonIQ2026.040.7570.815
Q-InsightCategory=MLLMs-based2026.010.7490.794
Qwen-SFTCategory=MLLM (w/ reasoning), Training Dataset=KonIQ2026.040.7390.813
DeQA-ScoreCategory=MLLMs-based2026.010.7380.743
Qwen2.5-VL-7BCategory=MLLMs-based2026.010.7350.772
Q-AlignCategory=MLLM-based, Training Dataset=KonIQ2025.100.7350.772
Q-AlignCategory=MLLM (w/o reasoning), Training Dataset=KonIQ2026.040.7350.772
DeQACategory=MLLM-based, Training Dataset=KonIQ2025.100.7290.809
DeQA-ScoreCategory=MLLM (w/o reasoning), Training Dataset=KonIQ2026.040.7290.809
ManIQACategory=Non-MLLM Deep-learning, Training Dataset=KonIQ2025.100.7230.685
UnifiedReward-TCategory=MLLMs-based2026.010.7220.745
RALICategory=Non-MLLM Deep-learning, Training Dataset=KonIQ2025.100.7150.779
LIQECategory=Non-MLLM Deep-learning, Training Dataset=KonIQ2025.100.6970.739
CLIP-IQA+Category=MLLM (w/o reasoning), Training Dataset=KonIQ2026.040.6850.736
Q-AlignCategory=MLLMs-based2026.010.6820.694
C2ScoreCategory=MLLM-based, Training Dataset=KonIQ2025.100.6710.777
C2ScoreCategory=MLLM (w/o reasoning), Training Dataset=KonIQ2026.040.6710.777
LIQECategory=MLLMs-based2026.010.6530.653
DBCNNCategory=Non-MLLM, Training Dataset=KonIQ2026.040.6410.73
HyperIQACategory=Non-MLLM, Training Dataset=KonIQ2026.040.640.702
CLIP-IQA+Category=Non-MLLM Deep-learning, Training Dataset=KonIQ2025.100.6360.736
ManIQACategory=Non-MLLM, Training Dataset=KonIQ2026.040.6360.723
MUSIQCategory=Non-MLLM Deep-learning, Training Dataset=KonIQ2025.100.630.722
MUSIQCategory=Non-MLLM, Training Dataset=KonIQ2026.040.630.722
UNIQUECategory=Deep Learning2026.010.6080.581
NIQECategory=Handcrafted2026.010.5330.56
NIQECategory=Handcrafted, Training Dataset=N/A2025.100.5330.56
BRISQUECategory=Handcrafted2026.010.4970.541
BRISQUECategory=Handcrafted, Training Dataset=N/A2025.100.4970.541
MUSIQCategory=Deep Learning2026.010.4940.434
MANIQACategory=Deep Learning2026.010.4220.448