No-Reference Image Quality Assessment on CSIQ 1.0 (test)
0.885PLCCQ-Tacit†
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Q-Tacit†Method Category=VLM-based, Training Protocol=Multi-task training set in [25] for GRPO calibration2026.03 | 0.885 | 0.85 | |
| Q-TacitMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.879 | 0.842 | |
| Q-Insight†Method Category=VLM-based, Training Protocol=Multi-task training set in [25] for GRPO calibration2026.03 | 0.87 | 0.824 | |
| DeQA-ScoreMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.787 | 0.744 | |
| CLIP-IQA+Method Category=Non-VLM Deep-learning, Training Protocol=Trained on KonIQ-10K2026.03 | 0.772 | 0.719 | |
| MUSIQMethod Category=Non-VLM Deep-learning, Training Protocol=Trained on KonIQ-10K2026.03 | 0.771 | 0.71 | |
| VisualQuality-R1Method Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.743 | 0.689 | |
| BRISQUEMethod Category=Handcrafted, Training Protocol=Trained on KonIQ-10K2026.03 | 0.74 | 0.556 | |
| Compare2ScoreMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.735 | 0.705 | |
| NIQEMethod Category=Handcrafted, Training Protocol=Trained on KonIQ-10K2026.03 | 0.718 | 0.628 | |
| NIMAMethod Category=Non-VLM Deep-learning, Training Protocol=Trained on KonIQ-10K2026.03 | 0.695 | 0.649 | |
| Q-InsightMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.685 | 0.64 | |
| Qwen-2.5-VL-7BMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.674 | 0.65 | |
| Q-AlignMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.671 | 0.737 | |
| ManIQAMethod Category=Non-VLM Deep-learning, Training Protocol=Trained on KonIQ-10K2026.03 | 0.623 | 0.627 |