MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving

TL;DR

MultiNet introduces a unified, real-time architecture for joint semantic classification, detection, and segmentation in autonomous driving, achieving high accuracy and efficiency on the KITTI dataset.

Contribution

It presents a simple, end-to-end trainable model that shares an encoder across tasks, enabling real-time performance for autonomous driving applications.

Findings

Outperforms state-of-the-art in road segmentation on KITTI

Processes all tasks in under 100 ms

Efficient joint reasoning for autonomous driving

Abstract

While most approaches to semantic reasoning have focused on improving performance, in this paper we argue that computational times are very important in order to enable real time applications such as autonomous driving. Towards this goal, we present an approach to joint classification, detection and semantic segmentation via a unified architecture where the encoder is shared amongst the three tasks. Our approach is very simple, can be trained end-to-end and performs extremely well in the challenging KITTI dataset, outperforming the state-of-the-art in the road segmentation task. Our approach is also very efficient, taking less than 100 ms to perform all tasks.