Few Shot Image Classification On Imagenet Fs 6
Metrics
Top-5 Accuracy (%)
Results
Performance results of various models on this benchmark
Model Name | Top-5 Accuracy (%) | Paper Title | Repository |
|---|---|---|---|
| KTCH (ResNet-50) | 77.6 | Few-shot Relational Reasoning via Connection Subgraph Pretraining | |
| KGTN-ens (ResNet-50, h+g, max) | 81.12 | KGTN-ens: Few-Shot Image Classification with Knowledge Graph Ensembles | |
| IdeMe-Net (ResNet-50) | 77.4 | Image Deformation Meta-Networks for One-Shot Learning | |
| SGM w/ G (ResNet-50) | 77.3 | Low-shot Visual Recognition by Shrinking and Hallucinating Features | |
| SGM + graph | 78.6 | Model-Agnostic Graph Regularization for Few-Shot Learning | - |
| LSD (ResNet-50) | 73.8 | Low-shot learning with large-scale diffusion | |
| SGM (ResNet-50) | 77.4 | Low-shot Visual Recognition by Shrinking and Hallucinating Features | |
| KGTN (ResNet-50) | 80.9 | Knowledge Graph Transfer Network for Few-Shot Recognition |
0 of 8 row(s) selected.