Paper Read - Simultaneous Object Detection and Semantic Segmentation

[NicolaBernini]

Overview

Reading Simultaneous Object Detection and Semantic Segmentation

Abstract

Both object detection in and semantic segmentation of camera images are important tasks for automated vehicles. Object detection is necessary so that the planning and behavior modules can reason about other road users. Semantic segmentation provides for example free space information and information about static and dynamic parts of the environment. There has been a lot of research to solve both tasks using Convolutional Neural Networks. These approaches give good results but are computationally demanding. In practice, a compromise has to be found between detection performance, detection quality and the number of tasks. Otherwise it is not possible to meet the real-time requirements of automated vehicles. In this work, we propose a neural network architecture to solve both tasks simultaneously. This architecture was designed to run with around 10 Hz on 1 MP images on current hardware. Our approach achieves a mean IoU of 61.2% for the semantic segmentation task on the challenging Cityscapes benchmark. It also achieves an average precision of 69.3% for cars and 67.7% on the moderate difficulty level of the KITTI benchmark.

[NicolaBernini]

Analysis

Identified key issue

• Autonomous driving requires perception
• Current Perception SOTA results seem to be promising for a product but SOTA is achieved outside of the robotic and business communities
• SOTA DNN are trained and tested in big server rooms with a lot of resources available hence as a result of this they are very big, resources hungry and rather slow (because of the high number of operations) to meet the real time constraints related to deployed robots and this is especially true in autonomous driving