AutoMR: A Universal Time Series Motion Recognition Pipeline

2025-02-21

AI Summary

This work addresses two key challenges in multimodal motion recognition: cross-dataset sensor format heterogeneity and time-consuming hyperparameter optimization. We propose the first end-to-end, fully automated, general-purpose framework for this task. The framework uniformly processes heterogeneous time-series sensor data, enabling fully automated preprocessing—including multimodal temporal alignment—model training, Optuna-based hyperparameter optimization, and standardized evaluation—without manual intervention. Its core innovations are zero-configuration cross-dataset transferability and plug-and-play edge deployment. Evaluated on ten diverse, heterogeneous datasets, the framework achieves state-of-the-art performance, improving average classification accuracy by 3.2% and accelerating end-to-end deployment throughput by 8× compared to conventional pipelines.

Technology Category

Intelligent Robots: Multimodal Perception & Sensor FusionComputer Vision: Multi-modal VisionMachine Learning: Multimodal Learning

Application Category

Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applicationsUser Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendationSearch and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

In this paper, we present an end-to-end automated motion recognition (AutoMR) pipeline designed for multimodal datasets. The proposed framework seamlessly integrates data preprocessing, model training, hyperparameter tuning, and evaluation, enabling robust performance across diverse scenarios. Our approach addresses two primary challenges: 1) variability in sensor data formats and parameters across datasets, which traditionally requires task-specific machine learning implementations, and 2) the complexity and time consumption of hyperparameter tuning for optimal model performance. Our library features an all-in-one solution incorporating QuartzNet as the core model, automated hyperparameter tuning, and comprehensive metrics tracking. Extensive experiments demonstrate its effectiveness on 10 diverse datasets, achieving state-of-the-art performance. This work lays a solid foundation for deploying motion-capture solutions across varied real-world applications.

Problem

• Automated motion recognition pipeline
• Integration of data preprocessing and model training
• Optimization of hyperparameter tuning complexity

Innovation

• Automated motion recognition pipeline
• Seamless integration of data processing
• Automated hyperparameter tuning with QuartzNet