Count Trackula

Count Trackula is an open-source iOS application that utilizes Apple's Vision framework to count people entering and exiting a designated space. Designed for museums, galleries, and exhibits, the app provides a lightweight, privacy-conscious solution for visitor analytics.

Read about our approach here.

Features

• Real-time visitor tracking: Uses the device camera to count people entering and exiting a space.
• Simultaneous counting: Can detect multiple people passing simultaneously.
• Privacy-focused: No images or personally identifiable information are stored or transmitted.
• Event-based reporting: Logs Enter and Exit events for tracking visitor movement.
• Firebase Firestore integration: Uses Boop for data collection, but can be configured for other backends.
• Optimized for fixed installations: Ideal for use in doorways, hallways, and other controlled entry points.

Use Case

Sticky Culture developed Count Trackula for internal use but we believe it can be valuable for other organizations looking to monitor visitor flow in a non-invasive way. This is particularly useful for:

• Small museums & galleries
• Temporary exhibitions
• Visitor experience research

Getting Started

To use Count Trackula on your iOS device, follow these steps:

1. Clone the Repository

   git clone https://github.com/StickyCulture/count-trackula.git
   cd count-trackula

2. Copy the Example.xcconfig and update the values

   cp Configuration/Example.xcconfig Configuration/Debug.xcconfig
   cp Configuration/Example.xcconfig Configuration/Release.xcconfig

3. Configure Firebase if using Firestore for data collection, otherwise you can modify the Analytics.swift file to implement your own backend

   • Create a Firebase project
   • Download GoogleService-Info.plist and place it in the Configuration/ directory

4. Build and Run Open the project in Xcode and run the app on an iOS device, such as iPhone.

Development Status

Please note that Count Trackula is currently a prototype in active development. While we use it internally, we do not assume highly accurate data at this stage. The app provides a general sense of visitation but we still have some kinks to figure out. Contributions are welcome as we continue to refine the application.