Research map / Image generation / Autoregressive and latent generation
Correct for Whom? Subjectivity and the Evaluation of Personalized Image Aesthetics Assessment Models
Samuel Goree, Weslie Khoo, David J. Crandall
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 |
|---|---|---|---|
| Use AADB aggregate scores to predict participants' pairwise preferences | AADB scores · PR-AADB participant image pairs | accuracy: 0.2875 to 0.575 | Mixed |
| Infer aggregate image scores from PR-AADB pairwise labels | PR-AADB inferred scores · PR-AADB participant image pairs | accuracy: 0.5 to 0.7875 | Mixed |
| Fine-tune predictions on 20 common image pairs per user (ablation) | Ren et al. personalized IAQA model · PR-AADB participant image pairs | mean accuracy | No clear effect |
| Fine-tune predictions on 20 common image pairs per user (ablation) | Ren et al. personalized IAQA model · PR-AADB participant image pairs | accuracy variance: 0.065 -> 0.129 | Mixed |
| Use demographics, aesthetic attributes, and image-content features to explain agreement with AADB scores | Logistic regression consistency analysis · PR-AADB / AADB label agreement | pseudo-R2: 2.3% | Hurt |
| Use AADB scores to predict individual preferences rather than evaluate within a specific user population | Personalized IAQA evaluation · PR-AADB participant image pairs | accuracy: 0.2875 to 0.575 | Mixed |
Closest papers by meaning
- Personalized Image Aesthetics (ICCV 2017 · similarity 0.92)
- Rethinking Personalized Aesthetics Assessment: Employing Physique Aesthetics Assessment as An Exemplification (CVPR 2025 · similarity 0.86)
- Advancing Comprehensive Aesthetic Insight with Multi-Scale Text-Guided Self-Supervised Learning (AAAI 2025 · similarity 0.86)
- Bridging Cognitive Gap: Hierarchical Description Learning for Artistic Image Aesthetics Assessment (AAAI 2026 · similarity 0.86)
- Learning Multi-Dimensional Human Preference for Text-to-Image Generation (CVPR 2024 · similarity 0.85)
- ArtiMuse: Fine-Grained Image Aesthetics Assessment with Joint Scoring and Expert-Level Understanding (CVPR 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.
