No-Reference Image Quality Assessment on KADID 1.0 (test)
0.766PLCCQ-Tacit†
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Q-Tacit†Method Category=VLM-based, Training Protocol=Multi-task training set in [25] for GRPO calibration2026.03 | 0.766 | 0.76 | |
| Q-TacitMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.754 | 0.749 | |
| Q-Insight†Method Category=VLM-based, Training Protocol=Multi-task training set in [25] for GRPO calibration2026.03 | 0.742 | 0.736 | |
| Q-InsightMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.702 | 0.702 | |
| VisualQuality-R1Method Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.701 | 0.694 | |
| DeQA-ScoreMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.694 | 0.687 | |
| Q-AlignMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.674 | 0.684 | |
| Qwen-2.5-VL-7BMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.668 | 0.663 | |
| CLIP-IQA+Method Category=Non-VLM Deep-learning, Training Protocol=Trained on KonIQ-10K2026.03 | 0.653 | 0.654 | |
| MUSIQMethod Category=Non-VLM Deep-learning, Training Protocol=Trained on KonIQ-10K2026.03 | 0.575 | 0.556 | |
| NIMAMethod Category=Non-VLM Deep-learning, Training Protocol=Trained on KonIQ-10K2026.03 | 0.532 | 0.535 | |
| Compare2ScoreMethod Category=VLM-based, Training Protocol=Trained on KonIQ-10K2026.03 | 0.5 | 0.453 | |
| ManIQAMethod Category=Non-VLM Deep-learning, Training Protocol=Trained on KonIQ-10K2026.03 | 0.499 | 0.465 | |
| NIQEMethod Category=Handcrafted, Training Protocol=Trained on KonIQ-10K2026.03 | 0.468 | 0.405 | |
| BRISQUEMethod Category=Handcrafted, Training Protocol=Trained on KonIQ-10K2026.03 | 0.429 | 0.356 |