Joint Estimation for Scanned Objects Using Iterative Correspondence

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional content digitizing systems face challenges in accurately capturing the structure and functionality of objects with movable components, particularly when dealing with noisy digital scans, as they require significant processing power and can only identify a limited type of joints with accuracy, leading to erroneous estimations.

Innovation Solution

The system estimates a joint of an object by using a plurality of digital scans, determining global transformations and joint parameters, and iteratively updating correspondences to select the most accurate joint, capable of handling noise and identifying a wider variety of joints through an iterative process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional scanning techniques are used to capture object geometry, then digital versions can be created quickly, but the functionality of movable components is not captured accurately

Engineering Contradiction:

Improvespeed of creating digital contentVSAvoidaccuracy of functionality capture

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system transitions from static scanning to dynamic scanning by capturing images of the object in multiple different poses or configurations. This allows the system to observe how components move relative to each other and automatically infer joint types and functionality without requiring manual animation or re-posing of the object.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system replaces manual animation and mechanical re-posing processes with automated image analysis algorithms. The computer system automatically detects correspondences between components across multiple images and computes joint parameters through mathematical optimization, eliminating the need for manual intervention.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If animations or large numbers of digital scans are used to reproduce object functionality, then accuracy improves, but processing power and time requirements increase significantly

Engineering Contradiction:

Improveaccuracy of joint estimationVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary correspondence matching between components across images before conducting the full joint estimation process. By pre-identifying which components correspond to each other and their relative transformations, the system reduces the computational complexity of the subsequent optimization steps.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system segments the object into multiple components and establishes correspondences between them independently. This allows the joint estimation problem to be broken down into smaller, more manageable sub-problems that can be solved more efficiently than treating the entire object as a single unit.

Inventive Principle:

Principle #1Segmentation

3Device complexity

If conventional systems perform sequential analysis across multiple digital scans, then processing is simpler, but estimation errors increase when digital scans contain noise

Engineering Contradiction:

Improvesimplicity of processing methodVSAvoidaccuracy with noisy scans

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The system merges information from multiple images and multiple correspondence sets into a unified joint estimation. By combining evidence from all available data simultaneously through joint optimization rather than sequential processing, the system achieves greater robustness to noise while maintaining computational feasibility.

Inventive Principle:

Principle #5Merging (Combining)

4Ease of operation

If conventional systems are used to identify object joints, then the process is straightforward, but the types of joints that can be identified are limited

Engineering Contradiction:

Improvesimplicity of joint identificationVSAvoidrange of joint types identifiable

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system implements a universal joint estimation framework that can handle multiple types of joints (hinge, slider, ball-and-socket, etc.) through a single unified mathematical formulation. The same correspondence-based optimization approach works for different joint types, making the system versatile without sacrificing simplicity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10380317B2Determining structure and functionality of scanned objects

Publication Date: 2019.08.13 ADOBE INC

AI summary

Methods and systems for generating digital models from objects. In particular, one or more embodiments determine a plurality of correspondences for first and second components of an object. One or more embodiments estimate a joint connecting the first and second components based on the correspondences. One or more embodiments jointly determine a global transformation and one or more joint parameters that map the plurality of components of the object from the first digital scan to the second digital scan. One or more embodiments also updating the correspondences based on the determined global transformation and parameter(s). One or more embodiments re-estimate the joint based on the updated correspondences. One or more embodiments select a candidate joint with a lowest error estimate from a plurality of candidate joints according to determined global transformations and joint parameter(s) for the candidate joints.