Crowdsourced Volumetric Video Capture for Missing Angle Coverage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Capturing high-quality volumetric video from dynamic subjects is challenging due to the difficulty in installing fixed cameras, especially in environments like parades, where mobile subjects require alternative methods for capturing video feeds at proper perspectives.

Innovation Solution

A crowdsourced volumetric video capture system that analyzes user-uploaded videos to identify closed loop contours, calculates weights for capture demand, and incentivizes users to capture missing angles and locations using geofences and weight-based incentives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If fixed cameras are installed to capture volumetric video, then capture quality and perspective control are improved, but installation difficulty and system complexity increase significantly in dynamic environments

Engineering Contradiction:
Improvecapture qualityVSAvoidinstallation difficulty
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system allows mobile devices to automatically capture and upload videos without requiring fixed camera installations. Users self-serve by capturing videos from their locations, and the system automatically processes these uploads to identify closed loop contours and determine additional capture needs, eliminating the need for complex fixed camera infrastructure.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system transitions from static fixed camera installations to dynamic mobile capture. The geofence and weight-based incentive system dynamically adapt to capture requirements, allowing the system to flexibly respond to changing capture needs based on real-time analysis of uploaded videos and identified contour completion status.

Inventive Principle:
Principle #15Dynamics

2Manufacturing precision

If multiple capture feeds are collected from various locations, then volumetric video coverage and completeness are improved, but capture time and coordination effort increase

Engineering Contradiction:
Improvecoverage completenessVSAvoidcapture time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The system implements a feedback loop where uploaded videos are automatically analyzed to identify closed loop contours and determine which contours are complete or incomplete. This feedback is communicated to users through the geofence incentive system, guiding them to capture specific missing angles and locations, thereby efficiently completing coverage without random or redundant captures.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary analysis of uploaded videos to identify closed loop contours and determine capture completeness before requesting additional captures. This preliminary action allows the system to precisely identify what is missing and guide subsequent capture efforts, avoiding unnecessary capture time and coordination.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If crowdsourced video feeds are utilized, then capture efficiency and resource utilization are improved, but video quality consistency and selection difficulty increase

Engineering Contradiction:
Improvecapture efficiencyVSAvoidselection difficulty
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system changes the parameter of video selection from manual quality assessment to automated contour-based evaluation. By analyzing whether videos contribute to completing closed loop contours, the system objectively selects valuable captures regardless of source, efficiently managing crowdsourced content without manual quality review while maintaining consistent selection criteria.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS12581052B2Crowdsourced, demand-based volumetric video creation
Publication Date: 2026.03.17 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US12581052B2 patent drawing
  • US12581052B2 patent drawing

AI summary

According to one embodiment, crowdsourced volumetric video capture is provided. The embodiment may include receiving a video captured by a user of a subject. The embodiment may also include creating a geofence based on a reference area surrounding the received video. The embodiment may further include identifying one or more locations and one or more capture angles needed to generate a volumetric video of the subject. The embodiment may also include calculating and assigning a weight to the one or more locations and/or one or more capture angles based on an importance in generation of the volumetric video. The embodiment may further include identifying one or more photographic capture devices within the geofence. The embodiment may also include presenting the one or more locations, one or more capture angles, and the weight assigned to each location and/or each capture angle to a user associated with the photographic capture devices.