Running Form Analysis App

Product Introduction

1. Definition: StrideIQ is a browser-based running form analysis application (technical category: computer vision-powered biomechanics tool) that evaluates gait patterns using local video processing.
2. Core Value Proposition: It enables runners to instantly self-diagnose common form flaws—like overstride, poor knee extension, or trunk lean—without coaching fees or lab equipment, prioritizing privacy and accessibility through client-side processing.

Main Features

1. Local Video Analysis Engine: Processes MP4/WebM videos entirely in-browser using TensorFlow.js pose detection. No data leaves the device, ensuring GDPR-compliant privacy while analyzing 5+ gait metrics (foot strike angle, leg retraction timing, etc.) via frame-by-frame skeleton tracking.
2. Dual-Precision Modeling: Offers "Lite" (real-time MobileNet model) for rapid feedback and "Full" (higher-accuracy MoveNet Thunder) modes. Lite prioritizes speed on mobile devices; Full enhances joint-angle measurement precision for deeper biomechanical assessment.
3. Structured Biomechanical Reporting: Generates quantified metrics (median values per stride) with visual overlays, flagging critical form deviations. Outputs include:
   • Form Flags (e.g., "Excessive Overstride: 15°+")
   • Key Metrics (contact time, stride length ratio)
   • Running economy optimization tips based on peer-reviewed kinesiology principles.

Problems Solved

1. Pain Point: Eliminates cost/time barriers to gait analysis. Traditional methods require $200+ motion-capture labs or coaching sessions; StrideIQ delivers instant feedback for $0.
2. Target Audience:
   • Self-coached recreational runners
   • Marathon trainers monitoring form fatigue
   • Physical therapists screening injury risks
   • Running clubs conducting group form workshops
3. Use Cases:
   • Pre-race form check using a smartphone video
   • Injury rehabilitation progress tracking
   • Technique refinement during tempo runs

Unique Advantages

1. Differentiation: Unlike subscription apps (e.g., Runners Connect) or wearables (Garmin Dynamics), StrideIQ requires no hardware, subscriptions, or data uploads. Competitors like Dartfish need manual annotation; this automates analysis in <2 minutes.
2. Key Innovation: Browser-native TensorFlow.js implementation reduces server dependency while handling 240fps slow-mo iPhone videos. Rule-based algorithms convert joint coordinates into actionable biomechanical insights—e.g., calculating overstride via ankle-knee-hip trigonometry.

Frequently Asked Questions (FAQ)

1. Does StrideIQ work on Android phones?
   Yes, but performance varies by device. High-end Androids/iPhones run Lite mode smoothly; budget devices may lag. For optimal speed, use desktops or iPad Pros.
2. How accurate is the running form analysis?
   Accuracy depends on video quality (1080p+/30fps minimum) and modeling: Lite mode detects obvious flaws (±5° margin); Full mode refines precision for metrics like foot strike angle. It identifies trends—not millimeter-perfect data.
3. Can I use StrideIQ for sprint analysis?
   Currently optimized for easy-to-tempo paces (4:00-7:00 min/km). Sprint mechanics require higher frame rates than most phones support; future updates may address this.
4. Why does my browser warn about high battery usage?
   Video decoding and pose estimation are computationally intensive. Chrome/Firefox may throttle performance on phones; close background apps and use AC power for longer analyses.