20230196718. IMAGE AUGMENTATION DEVICE AND METHOD simplified abstract (HON HAI PRECISION INDUSTRY CO., LTD.)

From WikiPatents
Jump to navigation Jump to search

IMAGE AUGMENTATION DEVICE AND METHOD

Organization Name

HON HAI PRECISION INDUSTRY CO., LTD.

Inventor(s)

Yung-Hui Li of New Taipei City (TW)

Shen-Hsuan Liu of New Taipei City (TW)

Wenny Ramadha Putri of New Taipei City (TW)

IMAGE AUGMENTATION DEVICE AND METHOD - A simplified explanation of the abstract

This abstract first appeared for US patent application 20230196718 titled 'IMAGE AUGMENTATION DEVICE AND METHOD

Simplified Explanation

The abstract describes an image augmentation device that uses a processor and memory to perform various operations. These operations include extracting object contours from image masks, superimposing the contours onto other image masks based on augmentation parameters, and generating sample images using a generative adversarial network model.

  • The device extracts object contours from image masks and superimposes them onto other image masks.
  • The augmentation parameters include contour scaling, moving distance, rotation angle, and a range for superimposition.
  • The device generates a third image mask that includes the first and second object contours.
  • A generative adversarial network model is used to generate sample images based on the object contours in the third image mask.
  • The device utilizes machine learning techniques for image generation and augmentation.

Potential Applications

  • Image editing and manipulation software
  • Virtual reality and augmented reality applications
  • Computer vision and object recognition systems
  • Training data generation for machine learning models

Problems Solved

  • Limited availability of training data for machine learning models
  • Difficulties in generating realistic and diverse sample images for training purposes
  • Challenges in accurately manipulating and augmenting object contours in images

Benefits

  • Improved accuracy and efficiency in generating sample images for machine learning models
  • Enhanced capabilities for image editing and manipulation
  • Increased realism and diversity in augmented reality and virtual reality applications
  • Facilitates the development and training of computer vision systems


Original Abstract Submitted

an image augmentation device is provided, which includes a memory and a processor. the processor is configured for performing following operations: extracting a first object contour from a first image mask, wherein the first object contour corresponds to a first label; superimposing the first object contour to a superimposed region in a second image mask according to an augmentation parameter to generate a third image mask, wherein the augmentation parameter includes a contour scaling parameter, a contour moving distance, a contour rotation angle and a range which can be superimposed, and the third image mask includes the first object contour and a second object contour in the second image mask; and generating a sample image which corresponds to according to the first object contour and the second object contour in the third image mask by a generative adversarial network model for performing machine learning.