Common questions and answers about Count-Cups.
Count-Cups is a cross-platform water intake tracker that uses computer vision to automatically detect and count your drinking gestures. It helps you stay hydrated by tracking your daily water consumption.
Count-Cups uses computer vision (OpenCV and optionally MediaPipe) to detect drinking gestures from your webcam. It analyzes face and hand positions to identify when you're drinking and automatically counts sips.
Yes, Count-Cups is free and open source under the MIT License.
Count-Cups supports Windows, macOS, and Linux.
Yes, Count-Cups requires a camera for automatic sip detection. You can also manually enter sips if detection isn't available.
See the Installation Guide for detailed instructions. You can either download pre-built binaries or install from source.
Minimum: Windows 10/macOS 10.14/Linux (Ubuntu 18.04+), Python 3.10+, 4GB RAM, 100MB storage, camera
Recommended: Windows 11/macOS 12+/Linux (Ubuntu 20.04+), Python 3.11+, 8GB RAM, 500MB storage, HD webcam
Only if installing from source. Pre-built binaries don't require Python.
Yes, you can install Count-Cups on multiple devices. Data is stored locally on each device.
Detection accuracy depends on lighting, camera position, and calibration. With proper setup, accuracy is typically 80-90%. You can manually add missed sips.
Yes, you can create multiple cup profiles with different sizes and types. Switch between profiles as needed.
Go to Settings and set your daily water intake goal in milliliters. The default is typically 2000ml (8 cups).
Yes, you can track any beverage. Create cup profiles for different beverages and Count-Cups will track them all.
Count-Cups needs to be running for detection. You can minimize to system tray on supported platforms.
Yes, all data is stored locally on your device. No data is transmitted to external servers unless you explicitly enable telemetry (opt-in).
No, Count-Cups does not record or store video. All processing happens in real-time and video frames are discarded immediately.
With telemetry disabled (default), no data is collected. If telemetry is enabled, only anonymous usage statistics are collected (no personal or hydration data).
Yes, you can export your data to CSV format from the History screen or Settings.
- Heuristics: Fast, default engine using OpenCV
- MediaPipe: More accurate, requires MediaPipe installation
Yes, you can adjust detection parameters in the Calibration screen and Settings.
Pre-built binaries: Download the latest release and replace the executable.
From source: git pull origin main && pip install -r requirements.txt --upgrade
Yes! See CONTRIBUTING.md for guidelines.
See Troubleshooting Guide for solutions. Common fixes: check permissions, try different camera index, update drivers.
Improve lighting, adjust camera position, calibrate settings, try different detection engine.
Check logs in ~/.count-cups/logs/, update dependencies, reset settings, report issue with log details.
Reduce camera resolution, close other applications, use heuristics engine, update graphics drivers.
This feature is planned for version 1.1. Currently, you can set one daily goal.
This feature is planned for version 1.3. Currently, data is stored locally on each device.
CSV export is available. Integration with health apps (Apple Health, Google Fit) is planned for future versions.
Mobile apps are planned for future versions. Currently, Count-Cups is desktop-only.
- Documentation: Check the documentation
- Troubleshooting: See Troubleshooting Guide
- GitHub Issues: Create an issue
- GitHub Discussions: Ask questions
- Email: contact@voxhash.dev
Create an issue on GitHub with:
- Description of the bug
- Steps to reproduce
- Expected vs actual behavior
- Environment information
- Log files
See SUPPORT.md for more details.
Create a feature request issue on GitHub with:
- Description of the feature
- Use case
- Proposed solution
- Alternatives considered
- Documentation: Browse the documentation index
- Troubleshooting: Check Troubleshooting Guide
- Support: Visit SUPPORT.md
Have a question not answered here? Create an issue on GitHub or contact us at contact@voxhash.dev.