arxiv_ml 90% Match Research Paper Performance Artists,HCI Researchers,Robotics Engineers,Interactive Media Designers 20 hours ago

Human-Machine Ritual: Synergic Performance through Real-Time Motion Recognition

robotics › human-robot-interaction

📄 Abstract

Abstract: We introduce a lightweight, real-time motion recognition system that enables synergic human-machine performance through wearable IMU sensor data, MiniRocket time-series classification, and responsive multimedia control. By mapping dancer-specific movement to sound through somatic memory and association, we propose an alternative approach to human-machine collaboration, one that preserves the expressive depth of the performing body while leveraging machine learning for attentive observation and responsiveness. We demonstrate that this human-centered design reliably supports high accuracy classification (<50 ms latency), offering a replicable framework to integrate dance-literate machines into creative, educational, and live performance contexts.

Key Contributions

Introduces a lightweight, real-time motion recognition system using IMU sensors and the MiniRocket algorithm for time-series classification. This system enables synergistic human-machine performance by mapping dancer-specific movements to sound and responsive multimedia control, preserving expressive depth while leveraging machine learning for attentive responsiveness.

Business Value

Opens new avenues for artistic expression and interactive experiences, enabling performers to collaborate dynamically with technology in live settings, potentially creating novel entertainment and educational tools.

Paper Metadata

Innovation Type

System Integration and Application

Deployment Feasibility

High. The system is described as lightweight, real-time, and uses readily available IMU sensors. The focus on dance suggests a specific but replicable application.

Limitations Addressed

Addresses the challenge of creating natural and expressive human-machine collaboration in performance contexts, moving beyond simple command-response interactions.

Technical Tags

Real-time Motion RecognitionIMU SensorsMiniRocketTime-Series ClassificationMultimedia ControlHuman-Machine CollaborationSomatic MemoryDance Performance

Research Topics

Human-Machine InteractionMotion RecognitionWearable ComputingCreative AIPerformance Art

Methods & Architectures

IMU Sensor Data ProcessingMiniRocket Time-Series ClassificationResponsive Multimedia Control MiniRocket

Applications & Tasks

Performing Arts Interactive Installations Wearable Technology Human-Computer Interaction Real-time motion trackingExpressive human-machine collaborationMapping movement to multimedia output Enabling synergistic performance between humans and machinesControlling multimedia elements in real-time based on human motion

Related Fields

Human-Computer InteractionRoboticsWearable ComputingSignal ProcessingMachine LearningPerforming Arts

Keywords

motion recognitionIMUMiniRockettime series classificationhuman-machine interactionsynergic performancewearable sensorsmultimedia controldancereal-time systemsexpressive performance

Academic Context

#Human-Machine Interaction#Motion Recognition#Wearable Computing#Creative AI#Performance Art

Technology Stack

Frameworks & Libraries

MiniRocket

Commercial Potential

Potential Products

Interactive performance systemsMotion-controlled art installationsWearable interfaces for creative expression

Target Industries

EntertainmentArtsEducationTechnology (Wearables)

Use Case Examples

A dancer controlling lighting and soundscapes with their movementsAn interactive museum exhibit that responds to visitor gesturesA system for choreographing performances with robotic elements

Competitive Edge

Offers a novel approach to human-machine collaboration in performance, focusing on expressive depth and real-time responsiveness using lightweight sensor technology.

Market Opportunity

Niche but growing market in interactive arts and performance technology.

Revenue Models

System salescustom installationslicensing for specific applications.

Resource Requirements

Compute Needs

Low, suitable for real-time processing on embedded systems or standard computers.

Data Requirements

Requires labeled IMU data for training the motion recognition model.

Deployment Constraints

Accuracy depends on sensor placement and calibration. Limited to movements detectable by IMUs.

Scalability

The classification algorithm (MiniRocket) is known for its efficiency, suggesting good scalability for real-time applications.

Production Readiness

Maturity Level

Prototype/Research

Time to Market

Short to Medium, depending on application development.

View Full Paper Back to Papers