TrapTracker RT — Real-Time Camera Trap Monitoring
Turn your camera traps into a live alerting system. TrapTracker RT is a Docker-based web application that watches an email inbox your cameras send to, identifies the species in each photo using AI, and emails your team the moment a species you care about appears.
It’s built for conservationists, ecologists and biodiversity practitioners who need to know what’s happening in the field now — not weeks later when the SD cards come back.
What it does
Most camera-trap workflows involve collecting cards and reviewing images long after the fact. TrapTracker RT closes that gap. A camera emails its photo, and the app:
- Downloads the image the moment it arrives
- Classifies the species with an ONNX AI model
- Checks the result against your project’s watch-list
- Emails every stakeholder the labelled image when a watched species is detected
Everything is organised into projects, each with its own cameras, mailbox, detection model, watch-list and recipients. You can run several projects at once, each monitoring a different inbox.
Key features
- Real-time email-driven pipeline — poll, detect and alert, automatically and continuously
- Per-project AI models — run UK mammals on one project and sub-Saharan African mammals on another; the app auto-discovers the models you provide
- Species watch-lists with confidence thresholds — alert only on what matters, at the certainty you choose
- Instant labelled-image alerts — recipients receive the annotated photo by email
- Multiple projects in parallel — monitor several inboxes and sites simultaneously
- Runs on any machine — CPU build works on Windows, macOS (including Apple Silicon) and Linux; optional NVIDIA GPU acceleration for higher throughput
- Your data stays yours — everything runs locally in Docker; nothing is sent anywhere except the alert emails you configure
How it works
Each camera sends its photos to a mailbox, with the camera’s name in the subject line in the form (=CameraName=). TrapTracker RT polls that mailbox, matches each email to a camera you’ve registered, extracts the image, runs detection, and — if a watch-list species is found above its confidence threshold — emails your recipients the annotated result. Detections and images are stored so you can review them in the built-in gallery.
Runs on your own hardware
TrapTracker RT is deployed with Docker and runs entirely on your own machine or server. The default build is CPU-based and works on virtually any computer, no special hardware required. If you have an NVIDIA GPU, an optional GPU build adds acceleration.
Tested with NVIDIA RTX A6000 (driver 573.73, CUDA 12.8). Minimum for GPU: NVIDIA driver 525+ (CUDA 12.x), x86_64. The CPU build has no special requirements.
For privacy and security, the application runs locally and is not designed to be exposed to the public internet.
Free for conservation and research
TrapTracker RT is free to use for non-commercial conservation, ecological research, biodiversity monitoring, academic research and education.
Commercial use — paid consultancy, resale, hosted services, or integration into proprietary commercial platforms — requires prior written permission.
Get it on GitHub
The full source, Docker files and step-by-step tutorial are available on GitHub:
https://github.com/traptracker-ai/traptracker-realtime-camera
AI model weights are released separately through Zenodo so they can be cited with a DOI. Please select the appropriate model from the Downloads/Models directory for specific model information and download link.