17436493. ELECTRONIC DEVICE AND METHOD OF INFERRING OBJECT IN IMAGE simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)

ELECTRONIC DEVICE AND METHOD OF INFERRING OBJECT IN IMAGE

Organization Name

SAMSUNG ELECTRONICS CO., LTD.

Inventor(s)

Kyounghoon Kim of Suwon-sikr (KR)

Sihoon Song of Suwon-si (KR)

Sangbok Han of Suwon-si (KR)

ELECTRONIC DEVICE AND METHOD OF INFERRING OBJECT IN IMAGE - A simplified explanation of the abstract

This abstract first appeared for US patent application 17436493 titled 'ELECTRONIC DEVICE AND METHOD OF INFERRING OBJECT IN IMAGE

Simplified Explanation

The abstract describes a method and electronic device for using a convolutional neural network (CNN) model to infer an object in an image. Here is a simplified explanation of the abstract:

• The method involves identifying a region of interest in a first frame of a moving image and determining the object within that region by passing the frame through a sequence of convolutional layer groups in the CNN model.
• A second region of interest is identified in a subsequent frame, corresponding to the first region, and provided to the CNN model.
• The method obtains output data from a specific convolutional layer group and uses it to decide whether to identify a second object in the second region of interest.

Potential applications of this technology:

• Object recognition and tracking in video surveillance systems.
• Autonomous vehicles for identifying and tracking objects in real-time.
• Augmented reality applications for object detection and interaction.

Problems solved by this technology:

• Efficiently identifying and tracking objects in a sequence of frames in a moving image.
• Reducing the computational complexity of object inference in real-time applications.

Benefits of this technology:

• Improved accuracy and speed in object recognition and tracking.
• Enhanced capabilities for real-time applications such as autonomous vehicles and augmented reality.
• Potential for automation and efficiency in various industries relying on object detection and tracking.

Original Abstract Submitted

Provided are a method and electronic device for inferring an object in an image using a convolutional neural network (CNN) model. The method includes including: identifying a first region of interest in a first frame in the moving image, and a first object in the first region of interest, by providing the first frame to convolution layer groups sequentially connected in the CNN model, identifying a second region of interest in a second frame, the second region of interest corresponding to the first region of interest, and the second frame being after the first frame, providing the second region of interest to the CNN model, and obtaining first output data output from a first convolution layer group from among the convolution layer groups, and determining whether to identify a second object in the second region of interest by using a second convolution layer group, based on the first output data.