An Empirical Evaluation of Deep Learning on Highway Driving
Brody Huval1, Tao Wang1, Sameep Tandon1, Jeff Kiske1, Will Song1, Joel Pazhayampallil1, Mykhaylo Andriluka1, Pranav Rajpurkar1, Toki Migimatsu1, Royce Cheng-Yue2, Fernando Mujica3, Adam Coates4, Andrew Y. Ng1 1Stanford University 2Twitter 3Texas Instruments 4Baidu Research
Abstract
Numerous groups have applied a variety of deep learning techniques to computer vision problems in highway perception scenarios. In this paper, we presented a number of empirical evaluations of recent deep learning advances. Computer vision, combined with deep learning, has the potential to bring about a relatively inexpensive, robust solution to autonomous driving. To prepare deep learning for industry uptake and practical applications, neural networks will require large data sets that represent all possible driving environments and scenarios. We collect a large data set of highway data and apply deep learning and computer vision algorithms to problems such as car and lane detection. We show how existing convolutional neural networks (CNNs) can be used to perform lane and vehicle detection while running at frame rates required for a real-time system. Our results lend credence to the hypothesis that deep learning holds promise for autonomous driving.
中文速览
高速公路自动驾驶中的车道线和车辆检测，长期依赖激光雷达、雷达等昂贵传感器，而摄像头配合传统计算机视觉方案又因鲁棒性不足难以实用。这篇文章用深度学习（deep learning）和卷积神经网络（convolutional neural network, CNN）来替代手工设计的视觉算法，在自行采集的高速公路大规模数据集上同时完成车道线检测和车辆检测两项任务。研究者以 Overfeat 检测框架为基础，结合目标掩码回归等改进手段，使整套系统在笔记本 GPU 上就能以超过 10 帧/秒的速度实时运行，并且无需额外的雷达或运动模型辅助。实验结果表明，CNN 在高速公路感知场景中具备足够的准确性和实时性，为以摄像头为核心、低成本且鲁棒的自动驾驶方案提供了有力的可行性证据。
原文 arXiv:1504.01716;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1504.01716v3