Poolability and Transferability in CNN. A Thrifty Approach

Jonathan Kobold, Vincent Vigneron, Hichem Maaref, Elmar Lang, Ana Maria Tomé

Keywords: Deep Learning, Receptive Field, Semantic Segmentation

TL;DR: The transfer layer reduces the size of CNN models for semantic segmentation.

Track: full conference paper

Paper Type: methodological development

Abstract: The current trend in deep learning models for semantic segmentation are ever increasing model sizes. These large models need huge data-sets to be trained properly. However medical applications often offer only small data-sets available and require smaller models. A large part of these models' parameters is due to their multi-resolution approach for increasing the receptive field, i.e. alternating convolution and pooling layers for feature extraction. In this work an alternative parameter free approach is proposed to increase the receptive field. This significantly reduces the number of parameters needed in semantic segmentation models and allows them to be trained on smaller data-sets.