EfficientSeg: An Efficient Semantic Segmentation Network

14 Sep 2020  ·  Vahit Bugra Yesilkaynak, Yusuf H. Sahin, Gozde Unal ·

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% mIoU) using the same parameter count (51.5% mIoU). Our most successful model obtained 58.1% mIoU score and got the fourth place in semantic segmentation track of ECCV 2020 VIPriors challenge.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Semantic Segmentation Cityscapes VIPriors subset EfficientSeg mIoU 58.03 # 1
Accuracy 81.68 # 1

Methods