MixSize training makes ImageNet classifiers resilient to smaller test images, matching baseline top-1 accuracy at 160x160 with about half the inference compute, while optionally improving accuracy or training speed.
Batch renormalization: Towards reducing minibatch dependence in batch-normalized models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency
MixSize training makes ImageNet classifiers resilient to smaller test images, matching baseline top-1 accuracy at 160x160 with about half the inference compute, while optionally improving accuracy or training speed.