DeepDriving: Learning Affordance for Direct Perception in Autonomous Driving

TL;DR

This paper introduces a third paradigm for autonomous driving that uses deep learning to directly estimate key perception indicators from images, providing a compact scene representation for driving decisions.

Contribution

It proposes a novel direct perception approach that maps images to key affordance indicators, bridging the gap between mediated perception and behavior reflex methods.

Findings

The deep CNN accurately predicts affordance indicators from virtual driving data.

The approach generalizes well to real-world images, demonstrated on the KITTI dataset.

The method enables simple control for autonomous driving based on scene affordances.

Abstract

Today, there are two major paradigms for vision-based autonomous driving systems: mediated perception approaches that parse an entire scene to make a driving decision, and behavior reflex approaches that directly map an input image to a driving action by a regressor. In this paper, we propose a third paradigm: a direct perception approach to estimate the affordance for driving. We propose to map an input image to a small number of key perception indicators that directly relate to the affordance of a road/traffic state for driving. Our representation provides a set of compact yet complete descriptions of the scene to enable a simple controller to drive autonomously. Falling in between the two extremes of mediated perception and behavior reflex, we argue that our direct perception representation provides the right level of abstraction. To demonstrate this, we train a deep Convolutional Neural Network using recording from 12 hours of human driving in a video game and show that our model can work well to drive a car in a very diverse set of virtual environments. We also train a model for car distance estimation on the KITTI dataset. Results show that our direct perception approach can generalize well to real driving images. Source code and data are available on our project website.