18223473. LONG-RANGE 3D OBJECT DETECTION USING 2D BOUNDING BOXES simplified abstract (NVIDIA Corporation)
Contents
LONG-RANGE 3D OBJECT DETECTION USING 2D BOUNDING BOXES
Organization Name
Inventor(s)
Zetong Yang of Rockville MD (US)
Zhiding Yu of Cupertino CA (US)
Chris Choy of Los Angeles CA (US)
Anima Anandkumar of Pasadena CA (US)
Jose M. Alvarez Lopez of Mountain View CA (US)
LONG-RANGE 3D OBJECT DETECTION USING 2D BOUNDING BOXES - A simplified explanation of the abstract
This abstract first appeared for US patent application 18223473 titled 'LONG-RANGE 3D OBJECT DETECTION USING 2D BOUNDING BOXES
The abstract discusses a patent application related to 3D object detection, focusing on long-range detection capabilities beyond the limitations of current lidar technology.
- The innovation in this patent application involves a technique for 3D object detection that supports long-range detection, addressing the need for applications like autonomous driving that require detection beyond the lidar range.
- Current techniques for 3D object detection rely on machine learning processes trained on images with high-quality 3D information, typically obtained using lidar technology.
- However, the limited measurable range of lidar technology poses challenges for detecting 3D objects beyond that range, which is crucial for safety-critical applications like autonomous driving.
- This patent application aims to enhance 3D object detection by enabling long-range detection capabilities, improving the safety and efficiency of applications that rely on accurate object detection in 3D space.
Potential Applications:
- Autonomous driving systems
- Robotics
- Surveillance systems
- Augmented reality applications
- Industrial automation
Problems Solved:
- Overcoming the limitations of current lidar-based 3D object detection systems
- Enabling long-range detection capabilities for improved safety and efficiency in various applications
Benefits:
- Enhanced safety in autonomous driving applications
- Improved accuracy and efficiency in object detection tasks
- Expanded capabilities for robotics and surveillance systems
Commercial Applications:
- This technology could revolutionize the autonomous driving industry by providing more reliable and accurate long-range object detection capabilities, leading to safer and more efficient autonomous vehicles.
Questions about 3D Object Detection: 1. How does this patent application address the limitations of current lidar-based 3D object detection systems? 2. What potential impact could this technology have on the development of autonomous driving systems?
Original Abstract Submitted
3D object detection is a computer vision task that generally detects (e.g. classifies and localizes) objects in 3D space from the 2D images or videos that capture the objects. Current techniques used for 3D object detection rely on machine learning processes that learn to detect 3D objects from existing images annotated with high-quality 3D information including depth information generally obtained using lidar technology. However, due to lidar's limited measurable range, current machine learning solutions to 3D object detection do not support detection of 3D objects beyond the lidar range, which is needed for numerous applications, including autonomous driving applications where existing close or midrange 3D object detection does not always meet the safety-critical requirement of autonomous driving. The present disclosure provides for 3D object detection using a technique that supports long-range detection (i.e. detection beyond the lidar range).
