17972831. LEARNING METHOD AND SYSTEM FOR OBJECT TRACKING BASED ON HYBRID NEURAL NETWORK simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
LEARNING METHOD AND SYSTEM FOR OBJECT TRACKING BASED ON HYBRID NEURAL NETWORK
Organization Name
Inventor(s)
MOON HYUN Cha of SUWON-SI (KR)
LEARNING METHOD AND SYSTEM FOR OBJECT TRACKING BASED ON HYBRID NEURAL NETWORK - A simplified explanation of the abstract
This abstract first appeared for US patent application 17972831 titled 'LEARNING METHOD AND SYSTEM FOR OBJECT TRACKING BASED ON HYBRID NEURAL NETWORK
Simplified Explanation
The object tracking learning system described in the patent application includes several modules that work together to improve object tracking accuracy. Here is a simplified explanation of the abstract:
- The system includes a first neural network module that learns and expresses a first parameter for an input image, converting it from one type to another.
- A second neural network module removes and learns a connection of a part of a second parameter for the input image.
- A prediction module generates a prediction value for an object in the input image by combining the results learned by the first and second neural network modules.
- An optimization module updates the first and second parameters based on the prediction value.
Potential Applications
This technology has potential applications in various fields, including:
- Object tracking in surveillance systems: The system can accurately track objects in surveillance videos, improving security and monitoring capabilities.
- Autonomous vehicles: The technology can be used to track objects on the road, helping autonomous vehicles navigate safely and avoid collisions.
- Augmented reality: By accurately tracking objects in real-time, this system can enhance the user experience in augmented reality applications.
Problems Solved
The object tracking learning system addresses the following problems:
- Inaccurate object tracking: By using neural networks and learning algorithms, the system improves the accuracy of object tracking, reducing false positives and negatives.
- Complex parameter learning: The system simplifies the learning process by dividing it into two modules, making it easier to learn and express the parameters for object tracking.
Benefits
The use of this technology offers several benefits:
- Improved object tracking accuracy: By combining the results learned by two neural network modules, the system provides more accurate predictions for object tracking.
- Efficient learning process: The system optimizes the learning process by dividing it into two modules, allowing for more efficient parameter learning.
- Versatile applications: The technology can be applied to various fields, such as surveillance, autonomous vehicles, and augmented reality, enhancing their capabilities.
Original Abstract Submitted
An object tracking learning system includes a first neural network module that expresses and learns a first parameter for an input image from a first type to a second type and outputs the learned result as a first learning result, a second neural network module that removes and learns a connection of a part of a second parameter for the input image and outputs the learned result as a second learning result, a prediction module that generates a prediction value for an object of the input image from a summation result obtained by summing the first learning result and the second learning result, and an optimization module that updates the first parameter and the second parameter based on the prediction value.
