A data-driven approach in cricket to achieve highly accurate results using minimal data.

---

Project Features

• Pitch-Map

  

  Objective

  Determine pitch map of deliveries by just using the Umpire POV video feed

  All the visualizations I used in this are mostly images that I provide along with explanations and based visualizations. generate-pitch-map.ipynb is attached to the report.

  This is a demo Desmos simulation for the ball pitching and perspective analysis. Each ball can be roughly assumed to be a skewed sine line in 3D space. To check that live : Click Here

---

• Ultra-Edge

  

  Objective

  Noise Suppression: Use two microphones (one near the bat, one further away) to filter out crowd noise. Frequency Analysis: Using a frequency spectrum tool, identify spikes in bat-ball strike frequency band gaps. Detect spikes in specific frequency ranges that indicate bat-ball strikes.

---

Prerequisites

• Python 3.x
• Required Python packages (listed in requirements.txt)

Setup Instructions

1. Clone the repository:

   git clone https://github.com/sudo-boo/cric-ai
   cd cric-ai

2. Install dependencies:

   Make sure you have pip installed, then run:

   pip install -r requirements.txt