3D Model Search System Contextual Query Augmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current visual search systems lack the ability to effectively utilize information from three-dimensional (3D) object data models to enhance search query results, particularly in terms of environment, orientation, position, and identification of other objects, leading to suboptimal search outcomes.
Innovation Solution
A method and system that receive a search query, identify a corresponding 3D model from a database, and update the query with information about the object's environment, orientation, position, and other objects, using this data to provide more relevant and organized search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If visual search systems use traditional 2D image matching, then the search process is simple and fast, but the search results lack contextual information about environment, orientation, and position
Solution Approach 1:
The patent transitions from traditional 2D image search to 3D model-based search by introducing three-dimensional spatial dimensions. The system creates 3D models from multiple 2D images and performs searches in 3D space, enabling retrieval of contextual information including environment, orientation, and position that cannot be obtained through conventional 2D matching alone.
Solution Approach 2:
The system performs preliminary actions by pre-processing multiple 2D images to generate 3D models before the actual search query is executed. These 3D models are stored with associated contextual metadata (environment, orientation, position) in advance, so when a search is performed, the system can quickly retrieve comprehensive contextual information without performing complex real-time analysis.
2Measurement precision
If the system retrieves comprehensive 3D model data including environment and orientation, then search result relevance improves, but processing time increases
Solution Approach 1:
The system performs preliminary actions by pre-processing multiple 2D images to generate 3D models before the actual search query is executed. These 3D models are stored with associated contextual metadata (environment, orientation, position) in advance, so when a search is performed, the system can quickly retrieve comprehensive contextual information without performing complex real-time analysis.
Solution Approach 2:
The system creates simplified 3D model representations and metadata copies that can be quickly searched and retrieved. Instead of processing full-resolution 3D data during queries, the system uses pre-generated model copies with extracted key features and contextual attributes, enabling fast search operations while maintaining high result accuracy.
Data Source
AI summary
A server may receive a search query indicative of an object. The server may identify a 3D model from within a database of 3D object data models that corresponds to the object. The 3D model may be associated with information that pertains to the object. The server may determine an updated search query based on the information associated with the 3D model. The updated search query may be indicative of one or more of an environment of the object, an orientation of the object in the environment, a position of the object in the environment, or an identification of one or more other objects in the environment. The server may provide one or more search query results that include data pertaining to the object based on the updated search query.