Parallel 3D Model Generation via Cloud Offloading

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Client devices, especially those with low computation capability and battery constraints, face difficulties in generating three-dimensional models from large collections of images due to the computationally intensive process, which can take days or weeks and is often impossible on mobile devices.

Innovation Solution

Offloading the computation of three-dimensional models to a network-accessible computation service, such as cloud-based servers, where images are processed in parallel by multiple servers to create and store the models, reducing the burden on client devices and enabling faster processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If three-dimensional models are computed on client devices, then users can generate and upload models, but the low computation capability and battery constraints of client devices make the process impractical

Engineering Contradiction:

Improvemodel generation speedVSAvoidcomputation capability

Core Design Contradiction:

ProductivityVSPower

Solution Approach 1:

A server acts as an intermediary between the client device and the model generation process. The client uploads images to the server, and the server performs the computationally intensive three-dimensional model generation using its superior processing power and parallel computing capabilities, resolving the contradiction between client device limitations and the need for fast model generation

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The computation is moved from the client device dimension to the server dimension, utilizing the server's distributed computing resources and parallel processing architecture to achieve high-speed model generation without burdening the client device's computation capability or battery

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Loss of time

If three-dimensional models are computed on client devices, then models can be generated locally, but the process takes days or weeks which is unacceptable

Engineering Contradiction:

Improvemodel generation timeVSAvoidprocessing throughput

Core Design Contradiction:

Loss of timeVSProductivity

Solution Approach 1:

The server performs model generation in advance after receiving the image upload, completing the computationally intensive work before the user needs to view the model. This preliminary action on the server eliminates the lengthy waiting period that would occur if computation were performed locally on the client device

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The server divides the large collection of images into multiple groups and processes them in parallel using multiple computing threads or distributed workers, dramatically reducing the time required to generate the three-dimensional model compared to sequential processing on a client device

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS8941644B2Computing three-dimensional image models in parallel

Publication Date: 2015.01.27 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

This document describes techniques for creating three-dimensional models of image collections using a network-accessible (e.g., cloud-based) computation service, rather than a client computing device. In some instances described below, the computation service performs at least a portion of the operations for creating these models in parallel.