Nvidia corporation (20240202936). REAL-TIME OBJECT TRACKING USING MOTION AND VISUAL CHARACTERISTICS FOR INTELLIGENT VIDEO ANALYTICS SYSTEMS simplified abstract

REAL-TIME OBJECT TRACKING USING MOTION AND VISUAL CHARACTERISTICS FOR INTELLIGENT VIDEO ANALYTICS SYSTEMS

Organization Name

nvidia corporation

Inventor(s)

Joonhwa Shin of Santa Clara CA (US)

Fangyu Li of San Jose CA (US)

Hugo Maxence Verjus of Zurich (CH)

Zheng Liu of Los Altos CA (US)

Kaustubh Purandare of San Jose CA (US)

REAL-TIME OBJECT TRACKING USING MOTION AND VISUAL CHARACTERISTICS FOR INTELLIGENT VIDEO ANALYTICS SYSTEMS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240202936 titled 'REAL-TIME OBJECT TRACKING USING MOTION AND VISUAL CHARACTERISTICS FOR INTELLIGENT VIDEO ANALYTICS SYSTEMS

Simplified Explanation

The patent application describes a method to identify objects in images based on their visual appearance and motion similarity.

Key Features and Innovation

• Obtaining visual appearance descriptors of objects in images from different time periods.
• Calculating compound similarity metrics between objects based on visual appearance and motion similarity.
• Updating object identifiers based on the similarity metrics meeting a threshold value.

Potential Applications

This technology can be used in:

• Surveillance systems for tracking objects over time.
• Augmented reality applications for object recognition.
• Autonomous vehicles for identifying and tracking objects on the road.

Problems Solved

• Efficient object identification and tracking in changing environments.
• Improved accuracy in recognizing objects based on visual appearance and motion.

Benefits

• Enhanced object recognition capabilities.
• Increased efficiency in tracking objects across different time periods.
• Improved overall performance of computer vision systems.

Commercial Applications

• "Enhanced Object Recognition and Tracking System for Surveillance Applications"

Prior Art

Further research can be conducted in the field of computer vision and object recognition to explore similar methods and technologies.

Frequently Updated Research

Stay updated on advancements in computer vision, object recognition, and motion tracking technologies to enhance the capabilities of this system.

Questions about Object Recognition and Tracking

How does this technology improve object identification in dynamic environments?

This technology utilizes visual appearance and motion similarity metrics to accurately track objects across different time periods, enhancing object identification in changing environments.

What are the potential applications of this technology beyond surveillance systems?

Apart from surveillance systems, this technology can be applied in augmented reality, autonomous vehicles, and various other fields requiring object recognition and tracking capabilities.

Original Abstract Submitted

a first visual appearance descriptor associated with a first object in an environment is obtained based on a first set of images of a first time period. the first object is subsequently absent from the environment in a second set of images of a second time period. a second visual appearance descriptor associated with a second object is obtained based on a third set of images, of a third time period subsequent to the second time period. a compound similarity metric between the first and second objects is obtained in view of visual appearance similarity and motion similarity metrics. the visual appearance similarity metric corresponds to a degree of similarity between the first and second visual appearance descriptors. an identifier associated with the second object is updated to correspond to an identifier associated with the first object in response to determining that the compound similarity metric meets a threshold value.