Abstract
In object detection systems for autonomous driving, LIDAR sensors provide very useful information. However, problems occur because the object representation is greatly distorted by changes in distance. To solve this problem, we propose a LIDAR shape set that reconstructs the shape surrounding the object more clearly by using the LIDAR point information projected on the object. The LIDAR shape set restores object shape edges from a bird’s eye view by filtering LIDAR points projected on a 2D pixel-based front view. In this study, we use this shape set for two purposes. The first is to supplement the shape set with a LIDAR Feature map, and the second is to divide the entire shape set according to the gradient of the depth and density to create a 2D and 3D bounding box proposal for each object. We present a multimodal fusion framework that classifies objects and restores the 3D pose of each object using enhanced feature maps and shape-based proposals. The network structure consists of a VGG -based object classifier that receives multiple inputs and a LIDAR-based Region Proposal Networks (RPN) that identifies object poses. It works in a very intuitive and efficient manner and can be extended to other classes other than vehicles. Our research has outperformed object classification accuracy (Average Precision, AP) and 3D pose restoration accuracy (3D bounding box recall rate) based on the latest studies conducted with KITTI data sets.
Highlights
Driver assistive technology to be mounted on a commercially available vehicle is rapidly developing
The first is to supplement the shape set with a LIDAR Feature map, and the second is to divide the entire shape set according to the gradient of the depth and density to create a 2D and 3D bounding box proposal for each object
Our research has outperformed object classification accuracy (Average Precision, AP) and 3D pose restoration accuracy (3D bounding box recall rate) based on the latest studies conducted with KITTI data sets
Summary
Driver assistive technology to be mounted on a commercially available vehicle is rapidly developing. The autonomous driving field requires a high accuracy and computational speed because it needs to recognize surrounding objects in fast moving vehicles In this respect, the object detection method based on deep learning was a suitable solution and many studies have been carried out. Given that the LIDAR Shape set represents the outline of an object in a 3D space, 2D and 3D proposals can be generated without adding a separate proposal generation method by adding the aspect ratio information per class This makes it possible to perform far fewer classifications than the conventional method of generating and classifying proposals for the entire image region, and to find objects with a higher accuracy with the same number of proposals. The code will be submitted to the benchmark competition after optimization
Sign-in/Register to view highlights & summary 
Published Version (
Free)
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 call