Intel Distribution of OpenVINO Toolkit: A Case Study of Semantic Segmentation

We provide an overview of the Intel Distribution of OpenVINO toolkit. The application of the OpenVINO toolkit is represented on the case study of semantic segmentation of on-road images. We provide a step-by-step tutorial for the problem solution based on the Dilation10 model, trained on the Cityscapes dataset. The Inference Engine component of the OpenVINO toolkit is used to implement inference of deep model. We focus on synchronous inference mode. Comparison of on-road semantic segmentation models supported by the OpenVINO toolkit is provided. Performance analysis of the Dilation10 inference implemented using various deep learning tools is carried out.