Silhouette-Based Limb Finding with Dual Distance Fields

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Skeletal tracking techniques are unreliable for detecting user behaviors in non-standard poses, such as when users are contorted, near the floor, or in sideways positions relative to the camera.

Innovation Solution

A silhouette-based limb finder system that uses two distance fields to track user limbs, allowing for the detection of body movements in various poses by generating a silhouette from a camera image and determining paths from extremity points to locations within the silhouette.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If skeletal tracking techniques are used to detect user movements, then tracking accuracy is improved for standard poses, but reliability deteriorates for non-standard poses (contorted positions, near floor, sideways positions)

Engineering Contradiction:

Improvetracking accuracyVSAvoidreliability for non-standard poses

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent segments the user's body into multiple limb components (arms, legs, torso) and tracks each segment independently using silhouette analysis. By dividing the body tracking problem into separate limb detection tasks, the system can reliably detect each body part's position and orientation even in non-standard poses where skeletal tracking fails.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Instead of using skeletal tracking that assumes standard anatomical configurations, the patent inverts the approach by using silhouette-based limb detection that makes no assumptions about pose correctness. The system detects limbs directly from the silhouette boundary and internal structure, working equally well for contorted, sideways, or floor-near positions as for standard poses.

Inventive Principle:

Principle #13The other way round (Inversion)

2Reliability

If silhouette-based limb detection with distance fields is implemented, then reliability for non-standard poses is improved, but device complexity increases

Engineering Contradiction:

Improvereliability for non-standard posesVSAvoidcomplexity of distance field processing

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces complex skeletal modeling and joint detection mechanisms with a simpler silhouette-based distance field approach. Instead of using sophisticated mechanical-like skeletal tracking algorithms that fail for non-standard poses, the system uses straightforward distance calculations from silhouette boundaries and internal points, achieving reliable limb detection with reduced computational complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces distance fields as an intermediary representation between the raw silhouette image and the final limb detection results. The distance fields serve as a computational mediator that simplifies the detection process by providing graded distance information from boundaries and internal points, making it easier to identify limb structures in various poses without complex direct analysis.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12443267B2Silhouette-based limb finder determination

Publication Date: 2025.10.14 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

A silhouette-based limb finder may be used to detect limbs from a camera image. This limb determination may be used to control an application, such as a game, or a combination with other image processing. A first distance field indicating a distance from the edge of a silhouette in an image and a second distance field indicating distance from a location in the silhouette may be used to generate a path from an extremity point on the silhouette to the location. This path then may be used to determine a limb in the silhouette. This allows tracking of limbs even for hard to detect player poses.