Aesthetics Quality Assessment On Ava
评估指标
Accuracy
评测结果
各个模型在此基准测试上的表现结果
模型名称 | Accuracy | Paper Title | Repository |
|---|---|---|---|
| DMA-Net | 75.4% | Deep Multi-Patch Aggregation Network for Image Style, Aesthetics, and Quality Estimation | - |
| Hand-crafted features | 68.0% | - | - |
| MP_adam | 83.0% | Attention-based Multi-Patch Aggregation for Image Aesthetic Assessment | |
| ADB-CNN | 77.3% | Photo Aesthetics Ranking Network with Attributes and Content Adaptation | - |
| NIMA | 81.5% | NIMA: Neural Image Assessment | - |
| MNA-CNN | 77.4% | Composition-Preserving Deep Photo Aesthetics Assessment | |
| A-Lamp | 82.5% | A-Lamp: Adaptive Layout-Aware Multi-Patch Deep Convolutional Neural Network for Photo Aesthetic Assessment | - |
| Pool-3FC | 81.7% | Effective Aesthetics Prediction with Multi-level Spatially Pooled Features | - |
| MTRLCNN | 79.1% | Deep Aesthetic Quality Assessment with Semantic Information | - |
0 of 9 row(s) selected.