# How to Build 3D Models Using Pictures from Different Angles > This patent describes a system that creates a three-dimensional digital object by combining multiple flat images taken from different viewpoints, then lets a user fine-tune how those images wrap around the object. - **Patent:** US 12737973 - **Original title:** 3D model generation using multiple textures - **Granted:** 2026 - **Status:** Active - **Times cited:** 0 - **Field:** consumer_electronics, software, telecommunications, gaming, ai_ml ## What it does The system first takes several "textures" (pictures) of an object, each showing a different side or view. It then automatically creates a rough 3D shape of the object and tries to stick these pictures onto the right parts of that shape, which is called an "initial alignment" (Abstract). A user can then give "input" to adjust how these pictures are placed and stretched on the 3D model (Abstract). Finally, all these separate pictures are blended into one single, complete picture that covers the entire 3D model, defining how it looks from all angles. This finished 3D model, along with its combined picture, is then saved. For example, if you wanted to create a 3D model of a toy car, you would take pictures of its front, back, sides, and top. The system would then build a car shape and let you adjust how those photos fit onto the digital car. ## What it does NOT cover - Generating 3D models solely from a single photograph or texture. - Creating 3D models without needing multiple textures from different views of the object. - Systems that generate 3D models without any user input to refine texture placement or alignment. - Methods that create 3D models using only depth data, LiDAR scans, or other non-texture-based inputs. - Generating 3D models where the multiple textures are not combined into a single, unified texture. ## The clever bit The core idea is taking multiple flat images, each from a different view, and intelligently combining them into a single, seamless texture that accurately covers a 3D model, allowing for user-guided refinement to ensure visual quality. ## Real-world examples 1. Creating assets for video games 2. Building virtual objects for augmented reality apps 3. Generating digital twins of real-world items 4. Producing visual effects for movies 5. 3D scanning services for e-commerce ## Why it matters Creating realistic 3D models is crucial for many digital experiences, especially in extended reality (XR) like virtual and augmented reality. This method helps streamline the process of making digital objects look accurate and detailed by efficiently using multiple visual inputs. This is important for industries building virtual worlds, training simulations, or digital product prototypes. ## Frequently asked questions ### What does How to Build 3D Models Using Pictures from Different Angles cover? This patent describes a system that creates a three-dimensional digital object by combining multiple flat images taken from different viewpoints, then lets a user fine-tune how those images wrap around the object. ### When does this patent expire? This patent is expected to expire on September 15, 2046, when the invention enters the public domain. ### What problem does this patent solve? Creating realistic 3D models is crucial for many digital experiences, especially in extended reality (XR) like virtual and augmented reality. This method helps streamline the process of making digital objects look accurate and detailed by efficiently using multiple visual inputs. This is important for industries building virtual worlds, training simulations, or digital product prototypes. ### What does this patent NOT cover? Generating 3D models solely from a single photograph or texture. **Full plain-English explainer:** https://patentbrief.org/patent/us/12737973/3d-model-generation-using-multiple-textures **Original patent:** https://patents.google.com/patent/US12737973 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [How to Build a 3D Model That Separates Light from Objects](https://patentbrief.org/patent/us/12737963/systems-and-methods-for-reconstructing-scenes-to-disentangle-light-and-matter) — This patent describes a method to create a detailed 3D digital model of a real scene that can tell the difference between the light in the scene and the objects it illuminates, even transparent ones. - [How AI Transfers Visual Styles Between Images While Keeping Structure](https://patentbrief.org/patent/us/12282696/method-and-system-for-semantic-appearance-transfer-using-splicing-vit-features) — This patent describes an AI method to generate a new image that combines the structure of one input image with the visual style of another, semantically related image, using a pre-trained Vision Transformer model. - [How Computers Track the Position and Angle of Multiple Objects](https://patentbrief.org/patent/us/12211234/rapid-iterative-development) — A method for using multiple cameras to track the exact location and orientation of many similar objects in a 3D space, even when they overlap. - [How to Combine Wide-Angle LiDAR Depth with Standard Camera Images](https://patentbrief.org/patent/us/11543533/systems-and-methods-for-wide-angle-lidar-using-non-uniform-magnification-optics) — This patent describes a method for combining a wide-angle depth map from a LiDAR sensor, which has unevenly spaced pixels, with a standard camera image by first correcting the depth map's pixel distribution and then merging the two for a complete view. - [Upscaling Images with AI and Sub-Pixel Information](https://patentbrief.org/patent/us/12737843/upsampling-an-image-using-one-or-more-neural-networks) — This patent describes how artificial intelligence, specifically neural networks, can make low-resolution images look sharper by cleverly guessing new pixel colors based on tiny details between existing pixels.