Bag of Tricks for Image Classification with Convolutional Neural Networks
This work addresses the need for better documentation and empirical validation of training tricks in image classification, benefiting researchers and practitioners by providing a consolidated set of effective refinements, though it is incremental as it builds on existing methods rather than introducing new paradigms.
The paper tackles the problem of improving image classification accuracy by systematically evaluating and combining training procedure refinements, such as data augmentations and optimization methods, which are often underreported. It shows that these refinements significantly boost CNN models, for example raising ResNet-50's top-1 validation accuracy from 75.3% to 79.29% on ImageNet, and also enhance transfer learning in domains like object detection and semantic segmentation.
Much of the recent progress made in image classification research can be credited to training procedure refinements, such as changes in data augmentations and optimization methods. In the literature, however, most refinements are either briefly mentioned as implementation details or only visible in source code. In this paper, we will examine a collection of such refinements and empirically evaluate their impact on the final model accuracy through ablation study. We will show that, by combining these refinements together, we are able to improve various CNN models significantly. For example, we raise ResNet-50's top-1 validation accuracy from 75.3% to 79.29% on ImageNet. We will also demonstrate that improvement on image classification accuracy leads to better transfer learning performance in other application domains such as object detection and semantic segmentation.