18279717. TEXTURE COMPLETION simplified abstract (Microsoft Technology Licensing, LLC)
Contents
TEXTURE COMPLETION
Organization Name
Microsoft Technology Licensing, LLC
Inventor(s)
TEXTURE COMPLETION - A simplified explanation of the abstract
This abstract first appeared for US patent application 18279717 titled 'TEXTURE COMPLETION
Simplified Explanation
The patent application describes a solution for completing textures of an object by generating a complete texture map from a partial texture map, making predictions on inferred textures, generating images based on the texture map, and determining if the images are generated images.
- Texture completion solution:
- Generate complete texture map from partial texture map - Make predictions on inferred textures - Generate images based on texture map - Determine if images are generated images
Potential Applications
The technology could be applied in: - Computer graphics - Virtual reality - Augmented reality
Problems Solved
This technology addresses: - Incomplete textures in objects - Improving image generation accuracy
Benefits
The benefits of this technology include: - Enhanced visual quality - Improved realism in virtual environments
Potential Commercial Applications
The technology could be used in: - Video game development - Architectural visualization - Product design
Possible Prior Art
One possible prior art is the use of texture synthesis algorithms to complete missing textures in images.
=== What is the accuracy rate of the texture discrimination model in predicting inferred textures? The accuracy rate of the texture discrimination model in predicting inferred textures is not specified in the abstract.
=== How is the image discrimination model trained to determine if the images are generated images? The abstract does not provide details on how the image discrimination model is trained to determine if the images are generated images.
Original Abstract Submitted
According to implementations of the present disclosure, there is provided a solution for completing textures of an object. In this solution, a complete texture map of an object is generated from a partial texture map of the object according to a texture generation model. A first prediction on whether a texture of at least one block in the complete texture map is an inferred texture is determined according to a texture discrimination model. A second image of the object is generated based on the complete texture map. A second prediction on whether the first image and the second image are generated images is determined according to an image discrimination model. The texture generation model, the texture and image discrimination models are trained based on the first and second predictions.
