18323795. OBJECT DETECTION USING POLYGONS FOR AUTONOMOUS SYSTEMS AND APPLICATIONS simplified abstract (NVIDIA Corporation)
Contents
- 1 OBJECT DETECTION USING POLYGONS FOR AUTONOMOUS SYSTEMS AND APPLICATIONS
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 OBJECT DETECTION USING POLYGONS FOR AUTONOMOUS SYSTEMS AND APPLICATIONS - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Key Features and Innovation
- 1.6 Potential Applications
- 1.7 Problems Solved
- 1.8 Benefits
- 1.9 Commercial Applications
- 1.10 Questions about Object Detection and Localization Technology
- 1.11 Original Abstract Submitted
OBJECT DETECTION USING POLYGONS FOR AUTONOMOUS SYSTEMS AND APPLICATIONS
Organization Name
Inventor(s)
Yang Zheng of Thousand Oak CA (US)
Trung Pham of Santa Clara CA (US)
Minwoo Park of Saratoga CA (US)
OBJECT DETECTION USING POLYGONS FOR AUTONOMOUS SYSTEMS AND APPLICATIONS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18323795 titled 'OBJECT DETECTION USING POLYGONS FOR AUTONOMOUS SYSTEMS AND APPLICATIONS
Simplified Explanation
The patent application discusses how object detectors can regress bounding polygons for detected objects in systems like autonomous driving. It explains the process of determining regression data to create accurate bounding shapes for objects in a scene.
- Object detectors regress bounding polygons for detected objects
- Regression data includes angles and lengths of vectors for accurate object localization
- Object detector can be trained by deforming regressed polygons to match ground truth polygons
Key Features and Innovation
- Object detectors regress bounding polygons for accurate object localization
- Regression data includes angles and lengths of vectors for precise object representation
- Training process involves deforming regressed polygons to match ground truth polygons
Potential Applications
This technology can be applied in autonomous driving systems, surveillance systems, robotics, and augmented reality applications for accurate object detection and localization.
Problems Solved
- Provides accurate object awareness and identification in various systems
- Enables precise object avoidance and localization for improved safety and efficiency
Benefits
- Enhances object detection accuracy in complex scenes
- Improves object localization for better decision-making in autonomous systems
Commercial Applications
Title: Advanced Object Detection and Localization Technology This technology can be utilized in autonomous vehicles, security systems, industrial automation, and virtual reality applications for enhanced object detection and localization capabilities, leading to safer and more efficient operations.
Questions about Object Detection and Localization Technology
How does this technology improve object detection accuracy?
This technology improves object detection accuracy by regressing bounding polygons and utilizing regression data to accurately localize objects in a scene, leading to more precise identification and awareness.
What are the potential applications of this technology beyond autonomous driving systems?
This technology can be applied in surveillance systems, robotics, and augmented reality applications for accurate object detection and localization, enhancing safety and efficiency in various industries.
Original Abstract Submitted
In various examples, one or more object detectors may regress bounding polygons for detected objects in systems (e.g., autonomous or semi-autonomous driving systems and applications) that provide object awareness, object identification, object avoidance, and/or object localization. The object detector may determine regression data representing a regressed polygon associated with a given shape of a detected object represented by classification data determined from a scene. The object detector may determine regression data for different regressed angles between different pairs of successive vertices of the regressed polygon and regressed lengths of vectors from a regressed geometric center of the regressed polygon to vertices of the regressed polygon. The object detector may generate, based at least in part on the regression data, a bounding shape for a detected object in the scene. In some embodiments, the object detector may be trained by deforming a regressed polygon to match a ground truth polygon.
