Research Article
Loss Architecture Search for Few-Shot Object Recognition
Table 4
Mean accuracy (%) ± CIs (%) of 14 state-of-the-art methods and LAS on the mini-ImageNet dataset with Conv-4 as the backbone.
| Method | 5-way accuracy | 1-Shot | 5-Shot |
| | 43.40 ± 0.78 | 51.09 ± 0.71 | MN- | 43.56 ± 0.84 | 55.31 ± 0.73 | ML- | 43.44 ± 0.77 | 60.60 ± 0.71 | | 45.10 | 55.20 | | 48.70 ± 1.84 | 63.11 ± 0.92 | Meta- | 49.21 ± 0.96 | — | | 49.42 ± 0.78 | 68.20 ± 0.66 | | 49.02 ± 0.98 | 63.50 ± 0.84 | | 50.44 ± 0.82 | 65.32 ± 0.70 | Meta- | 50.47 ± 1.87 | 64.03 ± 0.94 | | 51.24 ± 0.74 | 71.02 ± 0.64 | | 51.77 ± 1.86 | 65.60 ± 0.93 | MM-Net | 52.74 ± 0.45 | 65.82 ± 0.37 | MM-Net | 53.37 ± 0.48 | 66.97 ± 0.35 | | 54.61 ± 0.80 | 71.21 ± 0.66 | LAS | 55.14 ± 0.46 | 71.92 ± 0.48 |
|
|
Results from [ 32]. Results from [ 20]. Results from [ 33]. Results from [ 14]. |