用集成的LibFewShot完成的关于WebCaricature漫画人脸数据集的分类_AIGC-.zip

# Reproducible configs and checkpoints This folder contains: + Reproducible config of Table.1 in the paper of `LibFewShot`. + Reproducible config of Table.2 in the paper of `LibFewShot`. ## Reproduction results on miniImageNet (Results may different from the paper. Here are up-to-date results with checkpoints and configs.) <table> <tr> <td rowspan="2">Method</td> <td rowspan="2">Embed.</td> <td colspan="2">5-way 1-shot</td> <td colspan="2">5-way 5-shot</td> </tr> <tr> <td>Reported</td> <td>Ours</td> <td>Reported</td> <td>Ours</td> </tr> <tr> <td rowspan="2"> <a href="./Baseline">Baseline</a></td> <td>Conv64F</td> <td> 42.11 </td> <td> 42.34 </td> <td> 62.53 </td> <td> 62.18 </td> </tr> <tr> <td>ResNet18</td> <td> 51.75 </td> <td> 51.18 </td> <td> 74.27 </td> <td> 74.06 </td> </tr> <tr> <td rowspan="2"><a href="./Baseline++">Baseline++</a></td> <td>Conv64F</td> <td> 48.24 </td> <td> 46.21 </td> <td> 66.43 </td> <td> 65.18 </td> </tr> <tr> <td>ResNet18</td> <td> 51.87 </td> <td> 53.60 </td> <td> 75.68 </td> <td> 73.63 </td> </tr> <tr> <td><a href="./RFS/">RFS-simple</a></td> <td>ResNet12</td> <td> 62.02 </td> <td> 62.80 </td> <td> 79.64 </td> <td> 79.57 </td> </tr> <tr> <td><a href="./RFS/">RFS-distill</a></td> <td>ResNet12</td> <td> 64.82 </td> <td> 63.44 </td> <td> 82.14 </td> <td> 80.17 </td> </tr> <tr> <td><a href="./SKD">SKD-GEN0</a></td> <td>ResNet12</td> <td> 65.93 </td> <td> 66.40 </td> <td> 83.15 </td> <td> 83.06 </td> </tr> <tr> <td><a href="./SKD">SKD-GEN1</a></td> <td>ResNet12</td> <td> 67.04 </td> <td> 67.35 </td> <td> 83.54 </td> <td> 80.30 </td> </tr> <tr> <td><a href="./NegCos">NegCos</a></td> <td>ResNet12</td> <td> 63.85 </td> <td> 63.28 </td> <td> 81.57 </td> <td> 81.24 </td> </tr> <tr> <td><a href="./MAML">MAML</a></td> <td>Conv32F</td> <td> 48.70 </td> <td> 47.41 </td> <td> 63.11 </td> <td> 65.24 </td> </tr> <tr> <td><a href="./Versa">Versa</a></td> <td>Conv64F†</td> <td> 53.40 </td> <td> 51.92 </td> <td> 67.37 </td> <td> 66.26 </td> </tr> <tr> <td rowspan="2"><a href="./R2D2">R2D2</a></td> <td>Conv64F</td> <td> 49.50 </td> <td> 47.57 </td> <td> 65.40 </td> <td> 66.68 </td> </tr> <tr> <td>Conv64F‡</td> <td> 51.80 </td> <td> 55.53 </td> <td> 68.40 </td> <td> 70.79 </td> </tr> <tr> <td><a href="./ANIL">ANIL</a></td> <td>Conv32F</td> <td> 46.70 </td> <td> 48.44 </td> <td> 61.50 </td> <td> 64.35 </td> </tr> <tr> <td rowspan="2"><a href="./BOIL/">BOIL</a></td> <td>Conv64F</td> <td> 49.61 </td> <td> 48.00 </td> <td> 66.45 </td> <td> 64.39 </td> </tr> <tr> <td>ResNet12**</td> <td> - </td> <td> 58.87 </td> <td> 71.30 </td> <td> 72.88 </td> </tr> <tr> <td><a href="./MTL">MTL</a></td> <td>ResNet12</td> <td> 60.20 </td> <td> 60.20 </td> <td> 74.30 </td> <td> 75.86 </td> </tr> <tr> <td><a href="./Proto/">ProtoNet†</a></td> <td>Conv64F</td> <td> 46.14 </td> <td> 46.30 </td> <td> 65.77 </td> <td> 66.24 </td> </tr> <tr> <td><a href="./RelationNet">RelationNet</a></td> <td>Conv64F</td> <td> 50.44 </td> <td> 51.75 </td> <td> 65.32 </td> <td> 66.77 </td> </tr> <tr> <td><a href="./CovaMNet">CovaMNet</a></td> <td>Conv64F</td> <td> 51.19 </td> <td> 53.36 </td> <td> 67.65 </td> <td> 68.17 </td> </tr> <tr> <td rowspan="2"><a href="./DN4">DN4</a></td> <td>Conv64F</td> <td> 51.24 </td> <td> 51.95 </td> <td> 71.02 </td> <td> 71.42 </td> </tr> <tr> <td>ResNet12†</td> <td> 54.37 </td> <td> 57.76 </td> <td> 74.44 </td> <td> 77.57 </td> </tr> <tr> <td><a href="./CAN">CAN</a></td> <td>ResNet12</td> <td> 63.85 </td> <td> 66.62 </td> <td> 79.44 </td> <td> 78.96 </td> </tr> <tr> <td><a href="./RENet/">RENet</a></td> <td>ResNet12</td> <td> 67.60 </td> <td> 66.83 </td> <td> 82.58 </td> <td> 82.13 </td> </tr> </table> ## The overview picture of the SOTAs ### Conv64F <table> <tr> <td rowspan="2">Method</td> <td rowspan="2">Venue</td> <td rowspan="2">Type</td> <td colspan="2">miniImageNet</td> <td colspan="2">tieredImageNet</td> </tr> <tr> <td>1-shot</td> <td>5-shot</td> <td>1-shot</td> <td>5-shot</td> </tr> <tr> <td><a href="./Baseline">Baseline</a></td> <td>ICLR’19</td> <td> Non-episodic</td> <td> 44.90 </td> <td> 63.96 </td> <td> 48.20 </td> <td> 68.96 </td> </tr> <tr> <td><a href="./Baseline++">Baseline++</a></td> <td>ICML’19</td> <td> Non-episodic</td> <td> 48.86 </td> <td> 63.29 </td> <td> 55.94 </td> <td> 73.80 </td> </tr> <tr> <td><a href="./RFS">RFS-simple</a></td> <td>ECCV’20</td> <td> Non-episodic</td> <td> 47.97 </td> <td> 65.88 </td> <td> 52.21 </td> <td> 71.82 </td> </tr> <tr> <td><a href="./SKD">SKD-GEN0</a></td> <td>BMVC'20</td> <td> Non-episodic</td> <td> 48.14 </td> <td> 66.36 </td> <td> 51.78 </td> <td> 70.65 </td> </tr> <tr> <td><a href="./NegCos">NegCos</a></td> <td>ECCV’20</td> <td> Non-episodic</td> <td> 47.34 </td> <td> 65.97 </td> <td> 51.21 </td> <td> 71.57 </td> </tr> <tr> <td><a href="./MAML">MAML</a></td> <td>ICML’17</td> <td>Meta</td> <td> 49.55 </td> <td> 64.92 </td> <td> 50.98 </td> <td> 67.12 </td> </tr> <tr> <td><a href="./Versa">Versa</a></td> <td>NeurIPS’18</td> <td>Meta</td> <td> 52.75 </td> <td> 67.40 </td> <td> 52.28 </td> <td> 69.41 </td> </tr> <tr> <td><a href="./R2D2">R2D2</a></td> <td>ICLR’19</td> <td>Meta</td> <td> 51.19 </td> <td> 67.29 </td> <td> 52.18 </td> <td> 69.19 </td> </tr> <tr> <td><a href="./LEO">LEO</a></td> <td>ICLR’19</td> <td>Meta</td> <td> 53.31 </td> <td> 67.47 </td> <td> 58.15 </td> <td> 74.21 </td> </tr> <tr> <td><a href="./MTL">MTL</a></td> <td>CVPR’19</td> <td>Meta</td> <td> 40.97 </td> <td> 57.12 </td> <td> 42.36 </td> <td> 64.87 </td> </tr> <tr> <td><a href="./ANIL">ANIL</a></td> <td>ICLR’20</td> <td>Meta</td> <td> 48.01 </td> <td> 63.88 </td> <td> 49.05 </td> <td> 66.32 </td> </tr> <tr> <td><a href="./BOIL">BOIL</a></td> <td>ICLR’21</td> <td>Meta</td> <td> 47.92 </td> <td> 64.39 </td> <td> 50.04 </td> <td> 65.51 </td> </tr> <tr> <td><a href="./ProtoNet">ProtoNet</a></td> <td>NeurIPS’17</td> <td>Metric</td> <td> 47.05 </td> <td> 68.56 </td> <td> 46.11 </td> <td> 70.07 </td> </tr> <tr> <td><a href="./RelationNet">RelationNet</a></td>