18187595. GENERATING SEMANTICALLY-LABELLED THREE-DIMENSIONAL MODELS simplified abstract (QUALCOMM Incorporated)
GENERATING SEMANTICALLY-LABELLED THREE-DIMENSIONAL MODELS
Organization Name
Inventor(s)
Michel Adib Sarkis of San Diego CA (US)
GENERATING SEMANTICALLY-LABELLED THREE-DIMENSIONAL MODELS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18187595 titled 'GENERATING SEMANTICALLY-LABELLED THREE-DIMENSIONAL MODELS
- Simplified Explanation:**
This patent application describes a method for generating three-dimensional models using machine learning and segmentation masks.
- Key Features and Innovation:**
- Obtaining images of an object and segmentation masks with labels.
- Training a machine-learning model to create semantically-labeled 3D models.
- Generating a semantically-labeled 3D model of the object.
- Potential Applications:**
This technology can be used in various industries such as healthcare, gaming, virtual reality, and manufacturing for creating accurate 3D models of objects.
- Problems Solved:**
This technology addresses the challenge of efficiently generating detailed and labeled 3D models of objects using machine learning.
- Benefits:**
- Accurate and detailed 3D models can be generated.
- Automation of the modeling process saves time and resources.
- Improved understanding and visualization of objects in 3D space.
- Commercial Applications:**
Potential commercial applications include 3D modeling software, medical imaging technology, virtual reality development tools, and industrial design software.
- Questions about 3D Modeling:**
1. How does machine learning improve the process of generating 3D models? 2. What are the key benefits of using segmentation masks in creating 3D models?
- Frequently Updated Research:**
Research on improving the accuracy and efficiency of machine learning algorithms for generating 3D models is ongoing in the fields of computer vision and artificial intelligence.
Original Abstract Submitted
Systems and techniques are described herein for generating one or more three-dimensional models. For instance, a method for generating one or more three-dimensional models is provided. The method may include obtaining a plurality of images of an object; obtaining a plurality of segmentation masks associated with the plurality of images, each segmentation mask of the plurality of segmentation masks including at least one label indicative of at least one segment of the object in a respective image of the plurality of images; training, using the plurality of images and the plurality of segmentation masks, a machine-learning model to generate one or more semantically-labeled three-dimensional models of the object; and generating using the trained machine-learning model, a semantically-labeled three-dimensional model of the object, the semantically-labeled three-dimensional model of the object including at least one label indicative of the at least one segment of the object.
