Idea
A real-time motion tracking platform using smartphone and smartwatch sensors to enhance pose accuracy for fitness and health applications.
Research Paper
Core Innovation
This paper introduces BaroPoser, which uniquely integrates barometric data with IMU readings from everyday devices to improve human pose and translation estimation. It leverages sensor height changes and a novel local thigh coordinate frame to enhance accuracy on uneven terrain. This approach surpasses prior IMU-only methods in both benchmarks and real-world scenarios.
Market Size (TAM)
$2–10B TAM, $1–2B SAM; assumption: growing demand for wearable motion tracking in fitness, healthcare, and AR/VR sectors.
Potential Customers & Pain Points
- Fitness app developers needing accurate motion tracking on uneven terrain
- Healthcare providers requiring precise patient movement data
- AR/VR companies seeking improved human pose estimation
- Sports training platforms wanting real-time feedback
- Wearable device manufacturers aiming to enhance sensor fusion capabilities
Business Model
Licensing SDK/API to fitness, healthcare, and AR/VR companies; partnerships with wearable manufacturers; subscription for cloud analytics services.
Competitive Landscape
- Xsens
- Notch
- Perception Neuron
Implementation Challenges
- Integration complexity with diverse device sensors
- User privacy and data security concerns
- Competition from established motion tracking solutions
Validation Strategy
- Develop prototype SDK for smartphones and smartwatches
- Conduct pilot studies with fitness and healthcare partners
- Benchmark against leading IMU-only motion tracking solutions
Research Paper Overview
BaroPoser: Real-time Human Motion Tracking from IMUs and Barometers in Everyday Devices
Summary
BaroPoser combines IMU and barometric data from smartphones and smartwatches to estimate human pose and global translation in real time, improving accuracy on uneven terrain by leveraging sensor height changes and a novel local thigh coordinate frame. It outperforms state-of-the-art IMU-only methods on benchmark datasets and real-world recordings.