Cameras vs LiDAR Using Deep Learning

EasyChair Preprint 6293

Abstract

Research centers and companies dedicated to the development of autonomous vehicles are opting for two trends:using only cameras as a vision system or using LiDAR sensors plus cameras. Tesla has a fully camera vision system and Waymo has a cameras and LiDAR system. The difference of the LiDAR sensor could prevent accidents and save human lives in the future. Therefore, the main contribution of this work is the design of a methodology based on the comparison of detection efficiency for vision devices (Cameras and LiDAR). Applying measurement parameters provided by Neural Networks and models evaluation metrics in machine learning; it has to be concluded if its necessary to use LiDAR sensors in the development of autonomous cars.

Keyphrases: autonomous vehicles, neural networks, vision system

@booklet{EasyChair:6293,
  author    = {Fernando Rojas Ramos and Ivo Pineda Torres},
  title     = {Cameras vs LiDAR Using Deep Learning},
  howpublished = {EasyChair Preprint 6293},
  year      = {EasyChair, 2021}}