EfficientSeg: An Efficient Semantic Segmentation Network
share
Deep neural network training without pre-trained weights and few data is shown to need more training iterations. It is also known that, deeper models are more successful than their shallow counterparts for semantic segmentation task. Thus, we introduce EfficientSeg architecture, a modified and scalable version of U-Net, which can be efficiently trained despite its depth. We evaluated EfficientSeg architecture on Minicity dataset and outperformed U-Net baseline score (40 successful model obtained 58.1 segmentation track of ECCV 2020 VIPriors challenge.
READ FULL TEXT