17933378. OPTIMIZING COMPUTER-BASED GENERATION OF THREE-DIMENSIONAL VIRTUAL OBJECTS simplified abstract (International Business Machines Corporation)

OPTIMIZING COMPUTER-BASED GENERATION OF THREE-DIMENSIONAL VIRTUAL OBJECTS

Organization Name

International Business Machines Corporation

Inventor(s)

Jeremy R. Fox of Georgetown TX (US)

Martin G. Keen of Cary NC (US)

Alexander Reznicek of Troy NY (US)

Bahman Hekmatshoartabari of White Plains NY (US)

OPTIMIZING COMPUTER-BASED GENERATION OF THREE-DIMENSIONAL VIRTUAL OBJECTS - A simplified explanation of the abstract

This abstract first appeared for US patent application 17933378 titled 'OPTIMIZING COMPUTER-BASED GENERATION OF THREE-DIMENSIONAL VIRTUAL OBJECTS

Simplified Explanation

The present invention involves generating virtual objects in a three-dimensional virtual environment based on natural language utterances, using CLIP-guided Generative Latent Space analysis, monitoring usage of the generated virtual object, inferring human perception data, and generating a utility score based on the human perception data.

• Identification of virtual objects based on natural language utterances
• Generation of virtual objects using CLIP-guided Generative Latent Space analysis
• Monitoring of virtual object usage in the three-dimensional virtual space
• Inference of human perception data from monitoring
• Generation of utility score for virtual objects based on human perception data

Potential Applications

This technology could be applied in virtual reality gaming, virtual design and prototyping, virtual training simulations, and virtual art creation.

Problems Solved

This technology solves the problem of efficiently generating virtual objects in a three-dimensional virtual environment based on natural language input, while also taking into account human perception data for optimization.

Benefits

The benefits of this technology include improved user experience in virtual environments, enhanced customization and personalization of virtual objects, and more efficient creation of virtual content.

Potential Commercial Applications

A potential commercial application of this technology could be in the development of virtual reality software for various industries such as gaming, design, training, and entertainment.

Possible Prior Art

One possible prior art could be the use of natural language processing in virtual reality applications for generating virtual objects, but the specific combination of CLIP-guided Generative Latent Space analysis and human perception data inference may be novel.

How does this technology handle multi-user interactions in a virtual environment?

This article does not address how the technology deals with generating virtual objects for multiple users interacting in the same three-dimensional virtual space.

What are the potential privacy concerns related to monitoring human perception data in a virtual environment?

This article does not discuss the privacy implications of collecting and analyzing human perception data in the context of generating virtual objects.

Original Abstract Submitted

In an approach to improve the generation of a virtual object in a three-dimensional virtual environment, embodiments of the present invention identify a virtual object to be generated in a three-dimensional virtual environment based on a natural language utterance. Additionally, embodiments generate the virtual object based on a CLIP-guided Generative Latent Space (CLIP-GLS) analysis, and monitor usage of the generated virtual object in the three-dimensional virtual space. Moreover, embodiments infer human perception data from the monitoring, and generate a utility score for the virtual object based on the human perception data.