Abstract
Autonomous driving applications use two types of sensor systems to detect vehicles - depth sensing LiDAR and radiance sensing cameras. We compare the performance (average precision) of a ResNet for vehicle detection in complex, daytime, driving scenes when the input is a depth map [D = d(x,y)], a radiance image [L = r(x,y)], or both [D,L]. (1) When the spatial sampling resolution of the depth map and radiance image are both equal to typical camera resolutions, a ResNet detects vehicles at higher average precision from depth than radiance. (2) When the spatial sampling of the depth map matches the range of current LiDAR devices, the average precision is higher for radiance than depth. (3) A hybrid system that combines depth and radiance has higher average precision than systems using depth or radiance alone. We confirm these observations in both simulation and real-world data. We explain the advantage of combining depth and radiance by noting that the two types of information have complementary weaknesses. The radiance data are limited by dynamic range, motion blur and illumination variation; the LiDAR data have low spatial resolution. The ResNet effectively combines the two data sources to improve vehicle detection.
Highlights
People can detect vehicles using only one eye, and people with monocular vision are permitted to drive
EQUATED FOR SPATIAL RESOLUTION, DEPTH IS BETTER THAN RADIANCE FOR VEHICLE DETECTION IN COMPLEX SCENES How much information about the presence or absence of a vehicle is contained in high-quality depth measurements? We addressed this question by simulating pixel-wise depth maps from 3000 different driving scenes
The simulated depth maps have high spatial resolution, beyond what is typically measured by LiDAR and beyond the accuracy of explicit depth information that can be obtained from even the best stereo algorithms
Summary
People can detect vehicles using only one eye (monocular), and people with monocular vision are permitted to drive. People can accurately recognize objects from 2D images that contain no stereo information. These simple observations raise a practical question: Given the high accuracy of vehicle detection using monocular information or 2D images, how much will explicit depth information improve accuracy?. The second largest order of mammals after rodents, include many species that navigate through complex environments using depth sensing [14], [15]. There are several clear advantages to using depth sensing. Under low illumination conditions, such at night and in caves, radiance data are unreliable. Depth sensing avoids some of the challenging aspects of radiance measurements, such as non-uniform illumination and high dynamic range. How well can a system perform using only depth information?
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
