FMix: Enhancing Mixed Sample Data Augmentation

Mixed Sample Data Augmentation (MSDA) has received increasing attention in recent years, with many successful variants such as MixUp and CutMix. From insight on the efficacy of CutMix in particular, we propose FMix, an MSDA that uses binary masks obtained by applying a threshold to low frequency images sampled from Fourier space... (read more)

PDF Abstract ICLR 2021 PDF (under review) ICLR 2021 Abstract (under review)
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Image Classification CIFAR-10 PyramidNet + ShakeDrop + Fast AA + FMix Percentage correct 98.64 # 8
Image Classification CIFAR-100 DenseNet-BC-190 + FMix Percentage correct 83.95 # 20
Image Classification Fashion-MNIST PreAct-ResNet18 + FMix Percentage error 3.64 # 2

Methods used in the Paper