FLO — Fast Lock-On tracking for insect videography
FLO (Fast Lock-On) is free, open-source software for videography using a camera system that automatically moves to follow a subject. By continuously steering the camera to keep a fast-moving target centered in the field of view, FLO enables high-resolution outdoor videography of small, agile animals such as insects.
FLO receives image coordinates from Strand Camera and uses them to control motors that point the camera at the subject. It supports several hardware variants, from small servo-driven Mini-FLO setups to drone-based operation with MAVLink flight-controller integration and pilot video display.
FLO is described in the following publication:
Vo-Doan TT, Titov VV, Harrap MJM, Lochner S, Straw AD. High Resolution Outdoor Videography of Insects Using Fast Lock-On Tracking. Science Robotics 9(95), eadm7689 (2024). Free-access link DOI: 10.1126/scirobotics.adm7689.
The publication is accompanied by an open research dataset, analysis software, and build information for reproducing the tracking hardware.
Features
- Automatic, low-latency tracking that keeps a moving subject centered for high-resolution videography
- Builds on marker-free 2D tracking from Strand Camera
- Optional stereo cameras for distance estimation
- Multiple motor options, including PWM servo motors, Trinamic stepper motors, and SimpleBGC gimbal motors
- Drone-based operation with MAVLink flight-controller integration and on-screen display (OSD) for the pilot
- Web based user interface allowing easy remote control
- Built-in simulation tools to run and test the full controller with no camera or motor hardware attached
Source code and documentation
The primary repository, including source code, build instructions, and detailed documentation, is hosted on GitHub:
flo- the main FLO softwareflo-hardware- build instructions and manual for the BYO-camera tracking/filming setupflo-data-analysis- data analysis for FLO datared-button-trigger-timestamp- hardware to record times of external trigger events
Download
Source code is available via GitHub.
Licensing
FLO is free and open source software.
Support
Support is offered on a best effort basis via the github repository. Contact Andrew Straw regarding potential availability for collaboration or custom development.
For joint field-tracking or machine-learning projects, see the lab's collaboration interests.
Funding
FLO development was supported in part by the BeeTracked project, funded by the VolkswagenStiftung. Further support comes from the University of Freiburg.
