Research map / Image generation / Autoregressive and latent generation
Image Intrinsic Scale Assessment: Bridging the Gap Between Quality and Resolution
Vlad Hosu, Lorenzo Agnolucci, Daisuke Iso, Dietmar Saupe
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 |
|---|---|---|---|
| add WIISA weak labels | TOPIQ · IISA-DB test | SRCC: 0.764 -> 0.808 | Improved |
| add WIISA weak labels | DBCNN · IISA-DB test | SRCC: 0.755 -> 0.776 | Improved |
| add WIISA weak labels | CONTRIQUE · IISA-DB test | SRCC: 0.618 -> 0.631 | Improved |
| add WIISA weak labels | ARNIQA · IISA-DB test | SRCC: 0.651 -> 0.687 | Improved |
| add WIISA weak labels | CLIP-IQA+ · IISA-DB test | SRCC: 0.660 -> 0.666 | No clear effect |
| add WIISA weak labels | QualiCLIP · IISA-DB test | SRCC: 0.665 -> 0.670 | No clear effect |
| add WIISA weak labels | TOPIQ · IISA-DB test | RMSE: 0.098 -> 0.088 | Improved |
| use 2 weak labels instead of 1 (ablation) | WIISA TOPIQ · IISA-DB test | SRCC: 0.803 -> 0.808 | Improved |
| use 2 weak labels instead of 3 (ablation) | WIISA TOPIQ · IISA-DB test | SRCC: 0.788 -> 0.808 | Improved |
| use threshold delta 0.65 instead of 0.50 (ablation) | WIISA TOPIQ · IISA-DB test | SRCC: 0.795 -> 0.808 | Improved |
| use threshold delta 0.65 instead of 0.80 (ablation) | WIISA TOPIQ · IISA-DB test | SRCC: 0.802 -> 0.808 | Improved |
| use Lanczos instead of bilinear interpolation for weak labels (ablation) | WIISA TOPIQ · IISA-DB test | SRCC: 0.799 -> 0.808 | No clear effect |
Closest papers by meaning
- Adaptive Image Quality Assessment via Teaching Large Multimodal Model to Compare (NeurIPS 2024 · similarity 0.90)
- Toward Generalized Image Quality Assessment: Relaxing the Perfect Reference Quality Assumption (CVPR 2025 · similarity 0.89)
- Augmenting Perceptual Super-Resolution via Image Quality Predictors (CVPR 2025 · similarity 0.88)
- IQA-Spider: Unifying Multi-Granularity Image Quality Assessment with Reasoning, Grounding and Referring (ICML 2026 · similarity 0.86)
- Evaluation of Visual Processing Should Be Human-Centered, Not Metric-Centered (NeurIPS 2026 · similarity 0.86)
- FoMo: Forking Moment in Generative Trajectory as a Perceptual Distance (NeurIPS 2026 · similarity 0.85)
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.
