18435776. SYSTEM AND A METHOD FOR OBTAINING A PROCESSED OUTPUT IMAGE HAVING QUALITY INDEX SELECTABLE BY AN USER simplified abstract (Samsung Electronics Co., Ltd.)

From WikiPatents
Jump to navigation Jump to search

SYSTEM AND A METHOD FOR OBTAINING A PROCESSED OUTPUT IMAGE HAVING QUALITY INDEX SELECTABLE BY AN USER

Organization Name

Samsung Electronics Co., Ltd.

Inventor(s)

Aleksei Aleksandrovich Ivakhnenko of Moscow (RU)

Polina Vladimirovna Karpikova of Moscow (RU)

Anastasiia Sergeevna Iashchenko of Moscow (RU)

Andrei Nikolaevich Spiridonov of Moscow (RU)

Ekaterina Yurievna Radionova of Moscow (RU)

Riccardo Fabbricatore of Moscow (RU)

Leonid Igorevich Kostiushko of Moscow (RU)

SYSTEM AND A METHOD FOR OBTAINING A PROCESSED OUTPUT IMAGE HAVING QUALITY INDEX SELECTABLE BY AN USER - A simplified explanation of the abstract

This abstract first appeared for US patent application 18435776 titled 'SYSTEM AND A METHOD FOR OBTAINING A PROCESSED OUTPUT IMAGE HAVING QUALITY INDEX SELECTABLE BY AN USER

Simplified Explanation

The system described in the patent application allows users to select a generative artificial neural network (GAN) from a set stored in memory to obtain output images with a predefined quality index. Each pre-trained GAN includes multiple Earlier Exit branches, each containing calculating modules that perform the same function as corresponding modules in the backbone of the network.

  • Each GAN in the system is pre-trained and includes multiple Earlier Exit branches with calculating modules.
  • The user can select a GAN from the set stored in memory to generate images with a specific quality index.
  • The calculating modules in the Earlier Exit branches perform the same function as corresponding modules in the backbone of the network.

Potential Applications

This technology could be used in various fields such as image editing, graphic design, and digital art creation.

Problems Solved

This system allows users to easily generate high-quality images by selecting a specific GAN from a set of pre-trained networks.

Benefits

Users can obtain output images with a desired quality index by choosing a suitable GAN from the available options.

Potential Commercial Applications

This technology could be integrated into software applications for image editing, graphic design, and other creative industries.

Possible Prior Art

Prior art in the field of artificial neural networks and image generation techniques may include similar systems for generating images with predefined quality indexes.

What is the specific function of the Earlier Exit branches in the GANs described in the patent application?

The Earlier Exit branches in the GANs contain calculating modules that perform the same function as corresponding modules in the backbone of the network, allowing for the generation of high-quality images with predefined quality indexes.

How does the user select a specific GAN from the set stored in memory in the system described in the patent application?

The user can select a specific GAN by choosing from the set of pre-trained networks stored in memory, each offering different capabilities for generating images with varying quality indexes.


Original Abstract Submitted

A system for obtaining output images having a quality index selectable by a user includes an electronic device including at least one processor and a memory storing input images and a set of generative artificial neural networks (GANs), the at least one processor being configured to perform artificial neural network operations. Each GAN is selectable by the user from the set of GANs stored in the memory, for obtaining an image with predefined quality index, each GAN is pre-trained and includes a plurality of Earlier Exit branches each of which is connected after each calculating module, each Earlier Exit branch contains as many calculating modules as remain in the backbone from a connection point of that Earlier Exit branch to a backbone exit, and each calculating module of each Earlier Exit branch performs a same function as a corresponding remaining calculating module in the backbone.