17933902. SYSTEMS AND METHODS FOR MANIPULATED IMAGE DETECTION AND IMAGE RECONSTRUCTION simplified abstract (Verizon Patent and Licensing Inc.)
Contents
- 1 SYSTEMS AND METHODS FOR MANIPULATED IMAGE DETECTION AND IMAGE RECONSTRUCTION
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 SYSTEMS AND METHODS FOR MANIPULATED IMAGE DETECTION AND IMAGE RECONSTRUCTION - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
SYSTEMS AND METHODS FOR MANIPULATED IMAGE DETECTION AND IMAGE RECONSTRUCTION
Organization Name
Verizon Patent and Licensing Inc.
Inventor(s)
Saurabh Tahiliani of Noida (IN)
SYSTEMS AND METHODS FOR MANIPULATED IMAGE DETECTION AND IMAGE RECONSTRUCTION - A simplified explanation of the abstract
This abstract first appeared for US patent application 17933902 titled 'SYSTEMS AND METHODS FOR MANIPULATED IMAGE DETECTION AND IMAGE RECONSTRUCTION
Simplified Explanation
The method described in the abstract involves training a neural network to generate probable pixel values for masked portions of images, determining the contextual suitability of these values, and identifying unsuitable pixels in the images.
- Receiving a number of images to train a first neural network
- Masking a portion of each image and inputting the masked images to the first neural network
- Generating probable pixel values for the masked portions of the images
- Forwarding the images with probable pixel values to a second neural network
- Determining the contextual suitability of the probable pixel values
- Identifying pixels in the images that are not contextually suitable
Potential Applications
This technology could be applied in image editing software to automatically correct or enhance images by identifying and replacing pixels that are not contextually suitable.
Problems Solved
This technology addresses the challenge of accurately filling in missing or masked portions of images by using neural networks to generate and evaluate probable pixel values.
Benefits
The benefits of this technology include improved image editing capabilities, increased efficiency in image processing tasks, and enhanced image quality through the identification and correction of unsuitable pixels.
Potential Commercial Applications
A potential commercial application of this technology could be in the development of advanced image editing software for professionals in the photography and graphic design industries.
Possible Prior Art
One possible prior art for this technology could be the use of generative adversarial networks (GANs) in image inpainting tasks, where missing or corrupted portions of images are filled in using neural networks.
What are the specific neural network architectures used in this method?
The abstract does not provide details on the specific neural network architectures used in the method. It would be helpful to know the types of neural networks employed and how they are trained to perform the tasks described.
How does the second neural network determine the contextual suitability of the probable pixel values?
The abstract mentions that the second neural network determines whether each probable pixel value is contextually suitable, but it does not explain the specific criteria or mechanisms used for this evaluation. Understanding the process by which suitability is determined would provide more insight into the technology's functionality.
Original Abstract Submitted
A method may include receiving a number of images to train a first neural network, masking a portion of each of the images and inputting the masked images to the first neural network. The method may also include generating, by the first neural network, probable pixel values for pixels located in the masked portion of each of the plurality of images, forwarding the images including the probable pixel values to a second neural network and determining, by the second neural network, whether each of the probable pixel values is contextually suitable. The method may further include identifying pixels in each of the plurality of images that are not contextually suitable.
