Research map / Label-efficient and robust learning / Self-supervised and contrastive
Revisiting Image Aesthetic Assessment via Self-Supervised Feature Learning
Kekai Sheng, Weiming Dong, Menglei Chai, Guohui Wang, Peng Zhou, Feiyue Huang, Bao-Gang Hu, Rongrong Ji, Chongyang 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 |
|---|---|---|---|
| Jointly train degradation identification and triplet loss instead of using either alone (ablation) | Aesthetics-aware pretext task · AVA and AADB | accuracy: ∼ 0.3% on the AVA dataset and ∼ 0.7% on the AADB dataset | Improved |
| Add triplet loss to degradation identification training (ablation) | Aesthetics-aware pretext task · AVA and AADB | accuracy: ∼ 0.3% on the AVA dataset and ∼ 0.7% on the AADB dataset | Improved |
| Use triplet loss alone rather than the two-stage learning scheme (ablation) | Aesthetics-aware pretext task · AVA and AADB | training stability | Hurt |
| Add JPEG compression and Gaussian noise operations (ablation) | Proposed self-supervised pretraining · AVA and AADB | accuracy: ∼ 0.1% on AVA; ∼ 0.3% on AADB | Improved |
| Add exposure manipulation (ablation) | Proposed self-supervised pretraining · AVA and AADB | accuracy: ∼ 0.3% on AVA; ∼ 0.4% on AADB | Improved |
| Add downsampling, quantization, and pixelation operations (ablation) | Proposed self-supervised pretraining · AVA and AADB | accuracy: ∼ 0.4% on AVA; 0.5 ∼ 0.8% on AADB | Improved |
| Add Gaussian blur (ablation) | Proposed self-supervised pretraining · AVA and AADB | accuracy: ∼ 0.1% on AVA; ∼ 0.2% on AADB | Improved |
| Add rotation operation (ablation) | Proposed self-supervised pretraining · AVA and AADB | accuracy: 0.3 ∼ 0.5% on AVA; 0.4 ∼ 0.6% on AADB | Improved |
| Add patch shuffle and mixup operations (ablation) | Proposed self-supervised pretraining · AVA and AADB | accuracy: ∼ 0.2% on AVA; 0.3 ∼ 0.4% on AADB | Improved |
| Add entropy-based weights to patches (ablation) | Proposed self-supervised pretraining · Task adaptation benchmarks | accuracy: at least 0.5% | Improved |
| Use the proposed pretext task rather than ImageNet-label pretraining | Image aesthetic assessment · AVA, AADB, and CUHK-PQ | accuracy | Mixed |
Closest papers by meaning
- Advancing Comprehensive Aesthetic Insight with Multi-Scale Text-Guided Self-Supervised Learning (AAAI 2025 · similarity 0.92)
- Bridging Cognitive Gap: Hierarchical Description Learning for Artistic Image Aesthetics Assessment (AAAI 2026 · similarity 0.88)
- ELTA: An Enhancer against Long-Tail for Aesthetics-oriented Models (ICML 2024 · similarity 0.88)
- ArtiMuse: Fine-Grained Image Aesthetics Assessment with Joint Scoring and Expert-Level Understanding (CVPR 2026 · similarity 0.87)
- Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimization (ICLR 2026 · similarity 0.86)
- AACP: Aesthetics Assessment of Children’s Paintings Based on Self-Supervised Learning (AAAI 2024 · 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.
