Research map / Image generation / Autoregressive and latent generation
Active Fine-Tuning from gMAD Examples Improves Blind Image Quality Assessment
Zhihua Wang, Kede Ma
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| one round of active fine-tuning on gMAD set L1 | Baseline BIQA · gMAD set L1 | SRCC: 0.633 -> 0.818 | Improved |
| second round of active fine-tuning adding L2 | Baseline BIQA · gMAD set L3 | SRCC: 0.818 -> 0.813 | No clear effect |
| second round active fine-tuning | Baseline BIQA · Waterloo D-test | D-test: 0.890 -> 0.881 | No clear effect |
| one round of active fine-tuning | Baseline BIQA · Waterloo D-test | D-test: 0.881 -> 0.890 | Improved |
| active fine-tuning round 1 | Baseline BIQA · KonIQ-10k | SRCC: 0.403 -> 0.523 | Improved |
| active fine-tuning round 2 | Baseline BIQA · SPAQ | SRCC: 0.575 -> 0.633 | Improved |
| add gMAD examples versus simple fine-tuning on D2 only (round 2) (ablation) | Baseline BIQA · KADID-10k test subset | SRCC: 0.870 -> 0.872 | No clear effect |
| add gMAD examples versus simple fine-tuning on D2 only (round 2) (ablation) | Baseline BIQA · CSIQ test subset | SRCC: 0.851 -> 0.871 | Improved |
| add gMAD examples versus simple fine-tuning on D2 only (round 2) (ablation) | Baseline BIQA · TID2013 test subset | SRCC: 0.835 -> 0.828 | Hurt |
| fine-tune baseline on D2 after D1 pre-training | Baseline BIQA · TID2013 | SRCC: 0.675 -> 0.822 | Improved |
| fine-tune baseline on D2 after D1 pre-training | Baseline BIQA · LIVE | SRCC: 0.910 -> 0.896 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PieAPP — PieAPP: Perceptual Image-Error Assessment Through Pairwise Preference (CVPR 2018) · better on 2 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Towards Syn-to-Real IQA: A Novel Perspective on Reshaping Synthetic Data Distributions (NeurIPS 2025 · similarity 0.89)
- Toward Generalized Image Quality Assessment: Relaxing the Perfect Reference Quality Assumption (CVPR 2025 · similarity 0.88)
- Causal-IQA: Towards the Generalization of Image Quality Assessment Based on Causal Inference (ICML 2024 · similarity 0.87)
- Distilling Spatially-Heterogeneous Distortion Perception for Blind Image Quality Assessment (CVPR 2025 · similarity 0.87)
- Adaptive Image Quality Assessment via Teaching Large Multimodal Model to Compare (NeurIPS 2024 · similarity 0.87)
- Semi-Supervised Blind Image Quality Assessment through Knowledge Distillation and Incremental Learning (AAAI 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
