Long-tailed Image Classification on Places-LT
53.7Overall AccuracyLIFT w/ TTE
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| LIFT w/ TTEBackbone=ViT-L/14, Learnable Params.=0.27M, #Epochs=102023.09 | 53.7 | 53.1 | 54.1 | 53.8 | |
| LIFTBackbone=ViT-L/14, Learnable Params.=0.27M, #Epochs=102023.09 | 53.4 | 52.6 | 54.1 | 53.3 | |
| DecoderBackbone=ViT-L/14, Learnable Params.=39.79M, #Epochs=~342023.09 | 48.4 | — | — | — | |
| Zero-shot CLIPBackbone=ViT-L/14, Learnable Params.=-, #Epochs=-2023.09 | 39.9 | 38.1 | 39.2 | 45.1 |