Few-shot Classification on UCF101 16-shot (test)
75.33Accuracy (Sym 0.125)NA-MVP
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| NA-MVPImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 75.33 | 74.03 | 72.3 | 70.93 | 68.43 | 63.93 | 73.4 | 65.4 | |
| NLPromptImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 74.83 | 73.4 | 72.83 | 70.33 | 68.1 | 60.53 | 73.58 | 65.97 | |
| GCEImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 74 | 73.63 | 72.57 | 69.37 | 66 | 57.07 | 71.87 | 67.97 | |
| JoAPRImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 70.8 | 69.22 | 68.15 | 64.8 | 61.82 | 57.52 | 63.98 | 49.67 | |
| CoOpImage Encoder=ResNet-50, Text Encoder=63M-parameter Transformer, Training Shots=16, Training Epochs=502026.03 | 69.03 | 63.4 | 58.23 | 49.73 | 40.83 | 26.3 | 58.07 | 34.43 |