The headline in Gizmodo read “Elon Musk Was Right: Cheap Cameras Could Replace Lidar on Self-Driving Cars.” In a paper that will be presented at the 2019 Conference on Computer Vision and Pattern Recognition in June entitled, “Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving,” Cornell researchers detail a potential breakthrough for autonomous vehicles.
Cameras have typically been considered an inferior technology to Lidar given that they’re often installed at low angles, near a vehicle’s bumper, resulting in images that tend to distort objects in the distance which confuses neural networks trying to process and interpret the data.
But by placing a pair of cheap cameras on either side of a vehicle behind its windshield, stereoscopic images are produced which can be converted to 3D data. Because the images are being generated from a higher vantage point, closer to where Lidar systems are typically installed, the 3D data that was generated from the cameras was found to be nearly as precise as what laser scanners are able to generate, without distortion, and at a fraction of the cost.
Marta Hall, president of the San Jose-based Velodyne, said Musk’s comments are adding to confusion about autonomous vehicles: “I say this because of my concern for the roll out of the autonomous project without undue tragedy,” she said in a statement. “When it comes to Tesla’s claims, ‘consumer beware,’ is advised.”
Hall argues that Tesla and Musk have conflated being “almost autonomous” with becoming “fully autonomous.”
“Velodyne takes the position that ‘autonomy,’ implies that a car can drive safely and do all functions while a driver takes a nap,” she said. “Tesla offers what are really ‘driver features,’ not autonomy. They are cool features that are fun, but not for napping.”
And the beat goes on.
