Blind Image Quality Assessment on AGIQA (test)
0.816PLCCQ-Insight-Score
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Q-Insight-ScoreModel category=SFT-based and RL-based MLLMs, Input Condition=Image-conditioned2026.01 | 0.816 | 0.766 | |
| Q-Insight-ScoreCategory=SFT-based and RL-based MLLMs, Condition=Image-conditioned2025.12 | 0.816 | 0.766 | |
| Autoencoder-likeModel category=SFT-based and RL-based MLLMs, Input Condition=Image-conditioned2026.01 | 0.81 | 0.763 | |
| DeQAModel category=SFT-based and RL-based MLLMs, Input Condition=Image-conditioned2026.01 | 0.809 | 0.729 | |
| DeQACategory=SFT-based and RL-based MLLMs, Condition=Image-conditioned2025.12 | 0.809 | 0.729 | |
| Chain-of-ThoughtModel category=SFT-based and RL-based MLLMs, Input Condition=Image-conditioned2026.01 | 0.806 | 0.767 | |
| Self-ConsistencyModel category=SFT-based and RL-based MLLMs, Input Condition=Image-conditioned2026.01 | 0.805 | 0.766 | |
| OursCategory=SFT-based and RL-based MLLMs, Condition=Image-conditioned2025.12 | 0.803 | 0.76 | |
| Self-ConsistencyInput Condition=Text-Only2026.01 | 0.797 | 0.734 | |
| Self-ConsistencyInput Condition=Text-Only, Text Filtering=Score-related words removed2026.01 | 0.794 | 0.727 | |
| OursCategory=Caption-Only Conditions, Condition=Caption-Only2025.12 | 0.783 | 0.707 | |
| C2ScoreModel category=SFT-based and RL-based MLLMs, Input Condition=Image-conditioned2026.01 | 0.777 | 0.671 | |
| C2ScoreCategory=SFT-based and RL-based MLLMs, Condition=Image-conditioned2025.12 | 0.777 | 0.671 | |
| Autoencoder-likeInput Condition=Text-Only2026.01 | 0.774 | 0.696 | |
| Autoencoder-likeInput Condition=Text-Only, Text Filtering=Score-related words removed2026.01 | 0.774 | 0.696 | |
| Q-AlignModel category=SFT-based and RL-based MLLMs, Input Condition=Image-conditioned2026.01 | 0.772 | 0.735 | |
| Q-AlignCategory=SFT-based and RL-based MLLMs, Condition=Image-conditioned2025.12 | 0.772 | 0.735 | |
| Q-Insight-ScoreInput Condition=Text-Only2026.01 | 0.766 | 0.69 | |
| Chain-of-ThoughtInput Condition=Text-Only2026.01 | 0.766 | 0.69 | |
| Q-Insight-ScoreInput Condition=Text-Only, Text Filtering=Score-related words removed2026.01 | 0.766 | 0.689 | |
| Chain-of-ThoughtInput Condition=Text-Only, Text Filtering=Score-related words removed2026.01 | 0.766 | 0.689 | |
| CLIP-IQA+Model category=Deep-Learning Models, Input Condition=Image-conditioned2026.01 | 0.736 | 0.685 | |
| CLIP-IQA+Category=Deep-Learning Models, Condition=Image-conditioned2025.12 | 0.736 | 0.685 | |
| MANIQAModel category=Deep-Learning Models, Input Condition=Image-conditioned2026.01 | 0.723 | 0.636 | |
| MANIQACategory=Deep-Learning Models, Condition=Image-conditioned2025.12 | 0.723 | 0.636 | |
| MUSIQModel category=Deep-Learning Models, Input Condition=Image-conditioned2026.01 | 0.722 | 0.63 | |
| MUSIQCategory=Deep-Learning Models, Condition=Image-conditioned2025.12 | 0.722 | 0.63 | |
| NIMAModel category=Deep-Learning Models, Input Condition=Image-conditioned2026.01 | 0.715 | 0.654 | |
| NIMACategory=Deep-Learning Models, Condition=Image-conditioned2025.12 | 0.715 | 0.654 | |
| DBCNNModel category=Deep-Learning Models, Input Condition=Image-conditioned2026.01 | 0.641 | 0.648 | |
| DBCNNCategory=Deep-Learning Models, Condition=Image-conditioned2025.12 | 0.641 | 0.648 | |
| Q-Insight-ScoreCategory=Caption-Only Conditions, Condition=Caption-Only2025.12 | 0.639 | 0.592 | |
| NIQECategory=Hand-Crafted Models, Condition=Image-conditioned2025.12 | 0.56 | 0.533 | |
| BRISQUECategory=Hand-Crafted Models, Condition=Image-conditioned2025.12 | 0.541 | 0.497 |