3D Model Reconstruction Using Range Map Feature Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for forming 3D models from 2D images often result in incomplete models due to lack of information about the back side of objects, and combining range information from different viewpoints is challenging, especially in consumer applications where equipment is complex and expensive.
Innovation Solution
A method involving a data processing system that receives multiple images with corresponding range maps, identifies corresponding features, refines them using range maps, determines geometrical transformations, and aligns 3D images to create a complete 3D model, including all sides of an object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple images from different viewpoints are captured to obtain complete 3D model information, then the completeness of the 3D model is improved, but the complexity of combining and aligning range information from different viewpoints increases
Solution Approach 1:
The patent introduces a reference image as an intermediary to facilitate the alignment and combination of range information from multiple images. By designating one image as a reference and transforming other images to align with it, the system simplifies the complex task of merging multiple 3D datasets into a unified coordinate system, thereby reducing computational complexity while maintaining complete 3D model information.
2Device complexity
If conventional 2D stereo methods are used to determine range information, then the equipment complexity is reduced, but the measurement precision and reliability of depth information deteriorates
Solution Approach 1:
The patent combines multiple sources of information (2D image correspondence and range map data) to determine depth information. By merging the advantages of both conventional 2D stereo methods and range map techniques, the system achieves higher measurement precision while keeping equipment requirements relatively simple, as the combination of multiple information sources compensates for individual method limitations.
3Measurement precision
If structured lighting patterns are projected to obtain range information, then the measurement precision is improved, but the device complexity and cost increases
Solution Approach 1:
The patent makes the range information acquisition system multi-functional by using the same imaging device to capture both 2D images and range maps. This universal approach allows the system to perform multiple functions (2D imaging, depth mapping, and 3D reconstruction) with a single device, thereby improving measurement precision without proportionally increasing device complexity or cost.
4Ease of operation
If images are captured from a single viewpoint to simplify the imaging process, then the ease of operation is improved, but the completeness of the 3D model deteriorates due to missing back side information
Solution Approach 1:
The patent segments the 3D reconstruction process into multiple independent imaging steps from different viewpoints. By capturing images sequentially from multiple angles and then combining them through alignment and transformation operations, the system maintains operational simplicity for each individual capture while achieving complete 3D model information that includes previously hidden surfaces like back sides of objects.
Data Source
AI summary
A method for determining a three-dimensional model from three or more images comprising: receiving three or more images, each image being captured from a different viewpoint and including a two-dimensional image together with a corresponding range map; designating a plurality of pairs of received images, each pair including a first image and a second image. For each of the designated pairs a geometric transform is determined by identifying a set of corresponding features in the two-dimensional images; removing any extraneous corresponding features to produce a refined set of corresponding features; and determining the geometrical transformation for transforming three-dimensional coordinates for the first image to three-dimensional coordinates for the second image responsive to three-dimensional coordinates for the refined set of corresponding features. A three-dimensional model is determined responsive to the three or more received images and the geometrical transformations for the designated pairs of received images.