Nec corporation (20240096131). VIDEO PROCESSING SYSTEM, VIDEO PROCESSING METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM simplified abstract
Contents
- 1 VIDEO PROCESSING SYSTEM, VIDEO PROCESSING METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 VIDEO PROCESSING SYSTEM, VIDEO PROCESSING METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
VIDEO PROCESSING SYSTEM, VIDEO PROCESSING METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM
Organization Name
Inventor(s)
Daisuke Sugidomari of Tokyo (JP)
VIDEO PROCESSING SYSTEM, VIDEO PROCESSING METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240096131 titled 'VIDEO PROCESSING SYSTEM, VIDEO PROCESSING METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM
Simplified Explanation
The patent application describes a video processing system that includes various units to acquire, process, and store information related to individuals in video data.
- Image acquisition unit: Acquires at least one frame image from video data.
- Skeleton information generation unit: Generates skeleton information based on a person's body region in the frame image.
- Behavior conversion unit: Converts skeleton information into a behavior ID.
- Person specifying unit: Specifies a person ID based on the facial region of the person in the frame image.
- Registration unit: Registers behavior ID, person ID, and scene-related information in a database.
Potential Applications
This technology could be applied in security systems for identifying and tracking individuals, in sports analytics for analyzing player movements, and in healthcare for monitoring patient behavior.
Problems Solved
This technology solves the problem of efficiently identifying and tracking individuals in video data, as well as converting their movements into behavior IDs for analysis and storage.
Benefits
The benefits of this technology include improved security monitoring, enhanced sports performance analysis, and more accurate patient behavior tracking in healthcare settings.
Potential Commercial Applications
The technology could be commercially applied in surveillance systems, sports training facilities, and healthcare institutions for various monitoring and analysis purposes.
Possible Prior Art
One possible prior art could be existing video processing systems used in surveillance or sports analytics that track and analyze individual movements.
Unanswered Questions
The patent application does not provide details on how privacy concerns are addressed in the system. It would be important to understand the measures in place to protect individuals' privacy while using this technology.
What is the accuracy and reliability of the behavior conversion process in identifying and categorizing different behaviors?
The patent application does not mention the accuracy or reliability of the behavior conversion process. It would be crucial to know how well the system can accurately convert skeleton information into behavior IDs for practical applications.
Original Abstract Submitted
a video processing system () includes: an image acquisition unit () that acquires at least one frame image included in video data; a skeleton information generation unit () that generates skeleton information based on a body region of a person included in the at least one frame image; a behavior conversion unit () that converts the skeleton information into a behavior id; a person specifying unit () that specifies, based on a facial region of the person included in the at least one frame image, a person id for identifying features of a person estimated to be an identical person; and a registration unit () that registers the behavior id, the person id, and scene-related information related to the at least one frame image in a database in association with each other.
