Normalization in Training U-Net for 2D Biomedical Semantic Segmentation
This work addresses the problem of internal covariate shift in training deep convolutional neural networks for biomedical segmentation, which is incremental as it applies and compares existing normalization methods to a specific domain.
The paper compared four normalization methods (Batch, Instance, Layer, and Group Normalization) for training U-Net in 2D biomedical semantic segmentation, finding that Group Normalization with large groups or Instance Normalization achieved higher accuracy due to better model generalization, validated on three datasets (Right Ventricle, aorta, and Left Ventricle).
2D biomedical semantic segmentation is important for robotic vision in surgery. Segmentation methods based on Deep Convolutional Neural Network (DCNN) can out-perform conventional methods in terms of both accuracy and levels of automation. One common issue in training a DCNN for biomedical semantic segmentation is the internal covariate shift where the training of convolutional kernels is encumbered by the distribution change of input features, hence both the training speed and performance are decreased. Batch Normalization (BN) is the first proposed method for addressing internal covariate shift and is widely used. Instance Normalization (IN) and Layer Normalization (LN) have also been proposed. Group Normalization (GN) is proposed more recently and has not yet been applied to 2D biomedical semantic segmentation, however, no specific validations on GN were given. Most DCNNs for biomedical semantic segmentation adopt BN as the normalization method by default, without reviewing its performance. In this paper, four normalization methods - BN, IN, LN and GN are compared in details, specifically for 2D biomedical semantic segmentation. U-Net is adopted as the basic DCNN structure. Three datasets regarding the Right Ventricle (RV), aorta, and Left Ventricle (LV) are used for the validation. The results show that detailed subdivision of the feature map, i.e. GN with a large group number or IN, achieves higher accuracy. This accuracy improvement mainly comes from better model generalization. Codes are uploaded and maintained at Xiao-Yun Zhou's Github.