18562784. ADAPTIVE OBJECT DETECTION simplified abstract (MICROSOFT TECHNOLOGY LICENSING, LLC)
Contents
ADAPTIVE OBJECT DETECTION
Organization Name
MICROSOFT TECHNOLOGY LICENSING, LLC
Inventor(s)
Shiqi Jiang of Redmond WA (US)
Yuanchun Li of Redmond WA (US)
Yuanchao Shu of Redmond WA (US)
ADAPTIVE OBJECT DETECTION - A simplified explanation of the abstract
This abstract first appeared for US patent application 18562784 titled 'ADAPTIVE OBJECT DETECTION
Simplified Explanation: The patent application provides a solution for object detection by obtaining object distribution information and performance metrics. This information is used to generate detection plans for applying object detection models to target images captured by a camera, improving the balance between detection latency and accuracy.
- Object distribution information and performance metrics are obtained.
- Detection plans are generated based on this information.
- Plans indicate which object detection models to apply to target images.
- Improves balance between detection latency and accuracy.
Key Features and Innovation:
- Utilizes object distribution information and performance metrics for object detection.
- Generates detection plans based on the information obtained.
- Improves the balance between detection latency and accuracy in object detection.
Potential Applications: This technology can be applied in various fields such as surveillance, autonomous vehicles, and image recognition systems.
Problems Solved: This technology addresses the challenge of balancing detection latency and accuracy in object detection systems.
Benefits:
- Enhanced object detection performance.
- Improved efficiency in applying object detection models.
- Increased accuracy in detecting objects in images.
Commercial Applications: Potential commercial applications include security systems, traffic monitoring systems, and industrial automation processes.
Questions about Object Detection: 1. How does this technology improve the balance between detection latency and accuracy? 2. What are the potential applications of this object detection solution?
Frequently Updated Research: Stay updated on advancements in object detection technology and related research to enhance the performance of this solution.
Original Abstract Submitted
Implementations of the present disclosure provide a solution for object detection. In this solution, object distribution information and performance metrics are obtained. The object distribution information indicates a size distribution of detected objects in a set of historical images captured by a camera. The performance metric indicates corresponding performance levels of a set of predetermined object detection models. At least one detection plan is further generated based on the object distribution information and the performance metric. The at least one detection plan indicates which of the set of predetermined object detection models is to be applied to each of at least one sub-image in a target image to be captured by the camera. Additionally, the at least one detection plan is provided for object detection on the target image. In this way, a balance between the detection latency and the detection accuracy may be improved.
