Complementing Representation Deficiency in Few-shot Image Classification: A Meta-Learning Approach

Benchmark Model Rank Results
few-shot-image-classification-on-fc100-5-way-5-shotMCRNet-SVM#16Accuracy: 57.8
few-shot-image-classification-on-fc100-5-way-5-shotMCRNet-RR#19Accuracy: 56.6
few-shot-image-classification-on-fc100-5-way-1-shotMCRNet-SVM#18Accuracy: 41
few-shot-image-classification-on-fc100-5-way-1-shotMCRNet-RR#19Accuracy: 40.7
few-shot-image-classification-on-cifar-fs-5-way-5-shotMCRNet-SVM#23Accuracy: 86.8
few-shot-image-classification-on-cifar-fs-5-way-5-shotMCRNet-RR#28Accuracy: 85.2
few-shot-image-classification-on-cifar-fs-5-way-1-shotMCRNet-SVM#26Accuracy: 74.7
few-shot-image-classification-on-cifar-fs-5-way-1-shotMCRNet-RR#29Accuracy: 73.8
few-shot-image-classification-on-mini-imagenet-5-way-5-shotMCRNet-SVM#49Accuracy: 80.34
few-shot-image-classification-on-mini-imagenet-5-way-5-shotMCRNet-RR#56Accuracy: 78.16
few-shot-image-classification-on-mini-imagenet-5-way-1-shotMCRNet-SVM#62Accuracy: 62.53
few-shot-image-classification-on-mini-imagenet-5-way-1-shotMCRNet-RR#66Accuracy: 61.32