US Patent Application 18032490. SYNTHETIC DIGITAL IMAGE GENERATION simplified abstract

From WikiPatents
Jump to navigation Jump to search

SYNTHETIC DIGITAL IMAGE GENERATION

Organization Name

Telefonaktiebolaget LM Ericsson (publ)

Inventor(s)

Volodya Grancharov of Solna (SE)

Ludwig Thaung of Spanga (SE)

SYNTHETIC DIGITAL IMAGE GENERATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 18032490 titled 'SYNTHETIC DIGITAL IMAGE GENERATION

Simplified Explanation

The patent application describes a method for creating a synthetic digital image using an image processing device.

  • The method starts by obtaining an original digital image that represents a visual scene with objects.
  • The first step is to identify a specific object in the original image by using a bounding box.
  • A synthetic object is then generated by processing the first object and another object using a Generative Adversarial Network.
  • The synthetic object has a defined shape based on a binary segmentation mask applied to the first object.
  • The texture and color of the synthetic object are based on the second object.
  • Finally, the synthetic digital image is rendered by replacing the first object in the original image with the synthetic object.


Original Abstract Submitted

There is provided mechanisms for rendering a synthetic digital image. A method is performed by an image processing device. The method comprises obtaining (S) an original digital image. The original digital image represents a depiction of a visual scene that comprises at least one object. The method comprises identifying (S) a first object in the original digital image. The first object is bounded by a bounding box in the original digital image. The method comprises generating (S) a synthetic object by a Generative Adversarial Network processing the first object and a second object. The synthetic object has a shape defined by a binary segmentation mask as applied to the first object. The synthetic object has a texture and colour based on the second object. The method comprises rendering (S) the synthetic digital image by, in the original digital image, replacing the first object in the bounding box with the synthetic object.