Nvidia corporation (20240265555). OBJECT DETECTION USING IMAGE ALIGNMENT FOR AUTONOMOUS MACHINE APPLICATIONS simplified abstract

OBJECT DETECTION USING IMAGE ALIGNMENT FOR AUTONOMOUS MACHINE APPLICATIONS

Organization Name

nvidia corporation

Inventor(s)

Dong Zhang of Clarksville TN (US)

Sangmin Oh of San Jose CA (US)

Junghyun Kwon of Santa Clara CA (US)

Baris Evrim Demiroz of San Jose CA (US)

Tae Eun Choe of Belmont CA (US)

Minwoo Park of Saratoga CA (US)

Chethan Ningaraju of Munich (DE)

Hao Tsui of Munich (DE)

Eric Viscito of Shelburne VT (US)

Jagadeesh Sankaran of Dublin CA (US)

Yongqing Liang of San Jose CA (US)

OBJECT DETECTION USING IMAGE ALIGNMENT FOR AUTONOMOUS MACHINE APPLICATIONS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240265555 titled 'OBJECT DETECTION USING IMAGE ALIGNMENT FOR AUTONOMOUS MACHINE APPLICATIONS

Simplified Explanation: The patent application describes a system that uses a geometric approach to detect objects on a road surface by tracking points between frames and generating a disparity image to identify differences in location.

Key Features and Innovation:

• Geometric approach to object detection on road surfaces
• Tracking points between frames to determine location differences
• Generating a disparity image to identify objects
• Scoring and associating disparity values with pixels
• Generating a bounding shape based on scoring

Potential Applications: This technology could be used in autonomous vehicles, traffic monitoring systems, and surveillance cameras for object detection on road surfaces.

Problems Solved: This technology addresses the challenge of accurately detecting objects on road surfaces, which is crucial for road safety and traffic management.

Benefits:

• Improved object detection accuracy
• Enhanced road safety
• Efficient traffic management

Commercial Applications: Potential commercial applications include integration into autonomous vehicles, traffic monitoring systems for smart cities, and surveillance cameras for enhanced security measures.

Prior Art: Prior art related to this technology may include research on computer vision, object detection algorithms, and image processing techniques.

Frequently Updated Research: Researchers may be exploring advancements in computer vision algorithms, machine learning models for object detection, and real-time processing capabilities for road surface analysis.

Questions about Object Detection on Road Surfaces: 1. How does the geometric approach improve object detection accuracy compared to traditional methods? 2. What are the potential limitations of using a geometric approach for object detection on road surfaces?

Original Abstract Submitted

systems and methods are disclosed that use a geometric approach to detect objects on a road surface. a set of points within a region of interest between a first frame and a second frame are captured and tracked to determine a difference in location between the set of points in two frames. the first frame may be aligned with the second frame and the first pixel values of the first frame may be compared with the second pixel values of the second frame to generate a disparity image including third pixels. subsets of the third pixels that have an disparity image value about a first threshold may be combined, and the third pixels may be scored and associated with disparity values for each pixel of the one or more subsets of the third pixels. a bounding shape may be generated based on the scoring that corresponds to the object.