Research map / Image generation / Editing and style transfer
Towards Artistic Image Aesthetics Assessment: A Large-Scale Dataset and a New Method
Ran Yi, Haoyuan Tian, Zhihao Gu, Yu-Kun Lai, Paul L. Rosin
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 SAAN's style-specific and generic aesthetic branches together (ablation) | SAAN · BAID | SRCC: 0.425 -> 0.473 | Improved |
| Add the generic aesthetic branch (ablation) | SAAN · BAID | SRCC: 0.439 -> 0.473 | Improved |
| Add new global and art-related editing operations during pretraining (ablation) | SAAN · BAID | SRCC: 0.460 -> 0.473 | Improved |
| Use three distortion levels instead of two (ablation) | SAAN · BAID | SRCC: 0.462 -> 0.473 | Improved |
| Add spatial information fusion with a non-local block (ablation) | SAAN · BAID | SRCC: 0.459 -> 0.473 | Improved |
| Add cropping to pretraining operations (ablation) | SAAN · BAID | SRCC: 0.471 -> 0.473 | Improved |
| Add stylization to pretraining operations (ablation) | SAAN · BAID | SRCC: 0.471 -> 0.473 | Improved |
| Add convex distortion to pretraining operations (ablation) | SAAN · BAID | SRCC: 0.471 -> 0.473 | Improved |
| Add PencilSketch to pretraining operations (ablation) | SAAN · BAID | SRCC: 0.472 -> 0.473 | Improved |
| Add CutMix to pretraining operations (ablation) | SAAN · BAID | SRCC: 0.470 -> 0.473 | Improved |
| Use SAAN rather than TANet | BAID IAA assessment · BAID | SRCC: 0.453 -> 0.473 | Improved |
| Use SAAN rather than TANet | artistic IAA assessment · AVA | SRCC: 0.758 -> 0.742 | Hurt |
Closest papers by meaning
- ArtiMuse: Fine-Grained Image Aesthetics Assessment with Joint Scoring and Expert-Level Understanding (CVPR 2026 · similarity 0.91)
- Bridging Cognitive Gap: Hierarchical Description Learning for Artistic Image Aesthetics Assessment (AAAI 2026 · similarity 0.90)
- APDDv2: Aesthetics of Paintings and Drawings Dataset with Artist Labeled Scores and Comments (NeurIPS 2024 · similarity 0.90)
- Advancing Comprehensive Aesthetic Insight with Multi-Scale Text-Guided Self-Supervised Learning (AAAI 2025 · similarity 0.88)
- VADB: A Large-Scale Video Aesthetic Database with Professional and Multi-Dimensional Annotations (NeurIPS 2025 · similarity 0.87)
- ELTA: An Enhancer against Long-Tail for Aesthetics-oriented Models (ICML 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.
