Research map / Image generation / Autoregressive and latent generation
Photo Aesthetics Ranking Network with Attributes and Content Adaptation
Shu Kong, Xiaohui Shen, Zhe Lin, Radomir Mech, Charless Fowlkes
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 pairwise ranking loss to regression (ablation) | Reg · AADB test | Spearman rho: 0.6239 -> 0.6515 | Improved |
| add pairwise ranking loss to regression (ablation) | Reg · AVA test | Spearman rho: 0.4995 -> 0.5126 | Improved |
| add ranking loss to attribute-adapted model (ablation) | Reg+Att · AVA test | Spearman rho: 0.5331 -> 0.5445 | Improved |
| add ranking loss (ablation) | Reg · AVA test | accuracy (%): 72.04 -> 71.50 | No clear effect |
| sample pairs within-rater instead of cross-rater (ablation) | Reg+Rank · AADB test | Spearman rho: 0.6308 -> 0.6450 | Improved |
| combine within- and cross-rater sampling (ablation) | Reg+Rank · AADB test | Spearman rho: 0.6450 -> 0.6515 | Improved |
| sample 5 million instead of 2 million cross-rater pairs (ablation) | Reg+Rank · AADB test | Spearman rho: 0.6346 -> 0.6286 | No clear effect |
| sample 5 million instead of 2 million within-rater pairs (ablation) | Reg+Rank · AADB test | Spearman rho: 0.6450 -> 0.6448 | No clear effect |
| add attribute branch (ablation) | Reg+Rank · AVA test | Spearman rho: 0.5126 -> 0.5445 | Improved |
| add content-adaptive branches (ablation) | Reg+Rank · AVA test | Spearman rho: 0.5126 -> 0.5412 | Improved |
| add content-adaptive branches (ablation) | Reg+Rank · AADB test | Spearman rho: 0.6515 -> 0.6737 | Improved |
| add both attribute and content branches (ablation) | Reg+Rank · AVA test | accuracy (%): 71.50 -> 77.33 | Improved |
Closest papers by meaning
- Image Aesthetic Assessment Assisted by Attributes through Adversarial Learning (AAAI 2019 · similarity 0.92)
- Effective Aesthetics Prediction With Multi-Level Spatially Pooled Features (CVPR 2019 · similarity 0.90)
- Predicting Aesthetic Score Distribution Through Cumulative Jensen-Shannon Divergence (AAAI 2018 · similarity 0.90)
- Composition-Preserving Deep Photo Aesthetics Assessment (CVPR 2016 · similarity 0.89)
- Bridging Cognitive Gap: Hierarchical Description Learning for Artistic Image Aesthetics Assessment (AAAI 2026 · similarity 0.86)
- Advancing Comprehensive Aesthetic Insight with Multi-Scale Text-Guided Self-Supervised Learning (AAAI 2025 · similarity 0.86)
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.
