An efficient deep convolutional neural network for semantic segmentation

Owing to emergence of Deep Learning (DL) techniques, the segmentation models considerably enhanced their accuracy. However, this improved performance of currently popular DL models for self-driving car applications come at the cost of time and computational efficiency. Moreover, networks with efficient model architecture experience lack of accuracy. Therefore, in this study, we propose an efficient deep convolutional neural network (DCNN) model architecture that combines carefully formulated encoding and decoding paths. Specifically, the contraction path uses mixture of dilated and asymmetric convolution layers with skip connections and bottleneck layers, while the decoding path benefits from nearest neighbor interpolation method that demands no trainable parameters to restore original image size. Owing to the prudently designed model architecture, it considerably reduces the number of trainable parameters, required memory space, training, and inference time. Also, it obtained competitive accuracy scores on several evaluation metrics.