Dynamic Hand Gesture Region of Interest Localization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for detecting regions of interest in video frames using wearable devices are limited by their reliance on predefined shapes, are computationally expensive, and cannot handle dynamic hand gestures or objects beyond the field of view, making them unsuitable for wearable computing due to power and resource constraints.

Innovation Solution

A method that uses a hand segmentation algorithm to detect pixels corresponding to a fingertip in ego-centric video, computes the path of the fingertip through temporal analysis, and localizes the region of interest based on this path, allowing for dynamic hand gesture-based selection of regions of interest without pre-defined shapes, and performs actions on identified objects.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If predefined shape-based methods are used to detect regions of interest, then the method is simple to implement, but it cannot detect regions with unknown shapes and is computationally expensive

Engineering Contradiction:

Improveease of implementationVSAvoidadaptability to unknown shapes

Core Design Contradiction:

Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The patent transitions from static predefined shapes to dynamic hand gesture recognition. Instead of matching fixed templates, the system tracks the motion trajectory of the hand gesture across multiple video frames, allowing detection of any region shape that can be dynamically traced by a hand gesture.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent replaces the mechanical template matching approach with a computational vision-based approach. Instead of sliding windows over predefined shapes, the system uses hand gesture tracking algorithms that follow the user's hand motion to define regions of interest, significantly reducing computational complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If sliding-window-based template matching is used, then shape detection is accurate for predefined shapes, but power consumption is too high for wearable devices

Engineering Contradiction:

Improvedetection accuracyVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the essential information needed for region detection - the hand gesture trajectory - rather than processing every pixel and template combination. By focusing computation solely on tracking the hand's motion path across frames, the system achieves accurate detection with minimal computational overhead suitable for wearable devices.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses partial action by processing only the necessary portions of the video stream - specifically tracking the hand gesture trajectory rather than analyzing the entire frame. This selective processing maintains detection accuracy while dramatically reducing power consumption compared to full template matching.

Inventive Principle:

Principle #16Partial or excessive action

3Productivity

If static predefined shapes are used, then the method is computationally efficient, but it cannot handle dynamic hand gestures or regions beyond field of view

Engineering Contradiction:

Improvecomputational efficiencyVSAvoidhandling dynamic gestures

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamics by transitioning from static shape templates to dynamic hand gesture tracking. The system continuously updates the region of interest based on the hand's motion trajectory across multiple frames, enabling detection of dynamic gestures and regions that extend beyond the initial field of view as the user moves their hand.

Inventive Principle:

Principle #15Dynamics

4Reliability

If template matching is used, then the method works for known shapes, but it fails when the region of interest does not match predefined shapes

Engineering Contradiction:

Improvedetection reliability for known shapesVSAvoiddetection of unknown shapes

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent inverts the traditional approach by not starting with predefined shapes and checking if they match, but rather starting with the hand gesture trajectory and using it to define the region of interest. This inversion allows the system to adapt to any shape the user draws with their hand, making it universally applicable regardless of the region's geometry.

Inventive Principle:

Principle #13The other way round (Inversion)

Data Source

PatentUS9354711B2Dynamic hand-gesture-based region of interest localization

Publication Date: 2016.05.31 CONDUENT BUSINESS SERVICES LLC

AI summary

A method, non-transitory computer-readable medium, and apparatus for localizing a region of interest using a dynamic hand gesture are disclosed. For example, the method captures the ego-centric video containing the dynamic hand gesture, analyzes a frame of the ego-centric video to detect pixels that correspond to a fingertip using a hand segmentation algorithm, analyzes temporally one or more frames of the ego-centric video to compute a path of the fingertip in the dynamic hand gesture, localizes the region of interest based on the path of the fingertip in the dynamic hand gesture and performs an action based on an object in the region of interest.