⚡️ JumpPro: Biomechanical Video Analysis Lab

Professional Grade Performance Analysis in a Single HTML File.

🔴 Live Demo (在线试用) · 🛠️ Report Bug · 🚀 Request Feature

---

📖 Introduction (简介)

JumpPro is a client-side biomechanical analysis tool designed for S&C coaches and sports scientists. It runs entirely in your browser using a single HTML file.

Without expensive force plates, you can use high-frame-rate video (slow-motion) to accurately calculate key athletic metrics like Reactive Strength Index (RSI), Vertical Stiffness, and Power.

It includes an AI Coach (powered by DeepSeek) that analyzes the data against elite benchmarks and provides training advice.

JumpPro 是一个运行在浏览器端的轻量级运动生物力学分析工具。无需服务器，单文件即开即用。通过导入高帧率视频，即可精确计算 RSI、垂直刚度等核心指标，并内置 AI 教练提供专业建议。

✨ Features (核心功能)

• 🕵️‍♂️ Privacy First: Runs 100% locally in your browser. Video never uploads to any server.
• 📐 Scientific Algorithms:
  • RSI-mod (Countermovement Jump) based on flight time & takeoff time.
  • RSI (Drop Jump) based on contact time & jump height.
  • Vertical Stiffness ($K_{vert}$) using the Morin Sine-Wave Approximation Method.
  • Est. Peak Force using Samozino's method.
• 🎨 Silky Smooth UI: Frame-by-frame video scrubbing (Restored V8.5 Core), responsive design with Tailwind CSS & GSAP.
• 🤖 AI Coach Integration: Connects to DeepSeek API to generate professional training insights (NSCA-CSCS standard).
• 📂 Single File: Just download index.html and run.

🔬 The Science (科学原理)

JumpPro uses validated biomechanical formulas to estimate metrics from time-domain data:

1. Vertical Stiffness (Sine-Wave Method)

Based on the spring-mass model approximation (Morin et al.):

$$ K_{vert} = m \cdot \frac{\pi (t_f + t_c)}{t_c^2 \left( \frac{t_f + t_c}{\pi} - \frac{t_c}{4} \right)} $$

Where $t_f$ is flight time, $t_c$ is contact time, and $m$ is body mass.

2. Reactive Strength Index (RSI)

$$ \text{RSI} = \frac{\text{Jump Height (m)}}{\text{Contact Time (s)}} $$

🚀 Getting Started (如何使用)

Option A: Use Online (推荐)

Visit the Live Demo.

Option B: Run Locally

1. Download the index.html file from this repository.
2. Open it in Chrome, Safari, or Edge.
3. That's it!

📹 How to Analyze

1. Upload: Click "Import Video" (Use 60fps or 120fps/240fps slow-mo for best results).
2. Set Parameters: Enter athlete weight and video FPS in settings.
3. Mark Frames:
   • T1 (Start/Contact): Start of movement (CMJ) or Initial Contact (DJ). (Optional for CMJ)
   • T2 (Takeoff): The moment toes leave the ground.
   • T3 (Landing): The moment toes touch the ground again.
4. Analyze: Click the button to see metrics and AI feedback.

🛠 Tech Stack

• HTML5 Canvas for video rendering.
• Tailwind CSS (via CDN) for styling.
• GSAP for smooth animations.
• DeepSeek API for LLM integration.

🤝 Contributing

Pull requests are welcome! For major changes, please open an issue first to discuss what you would like to change.

📄 License

MIT © 2025 HarleyXu