Glad to announce the 2nd tutorial on Firetail 10 – ‘working with acceleration annotation’ featuring practical workflows for (1) acceleration annotation with FireSOM (2) ML model transfer across tags (3) overlay (video) annotation ground truth
Free trial: www.firetail.de
Our first tutorial on Firetail 10 is now available!
Density estimation on time ranges, featuring amazing data by Carlson et al. 2021
Movebank repository doi:10.1038/s41467-021-24826-x
We’re pleased to announce the launch of our Firetail Fellowship Program!
Take the opportunity!
Selected projects will receive a full Firetail license and benefit from our expert support.
Firetail 10 is now available at www.firetail.de
Rapid visualization and analysis of animal data. Get it today and view data in ways you never have before. Ideal for Movebank resources and high precision tags like e-obs, Technosmart, Ornitela, Vectronic, Milsar and many others.
Our contribution to the Wildlabs Variety hour is now available.
The full event can be found here:
We’re glad to join the Wildlabs Variety Hour: Firetail developer Tobias Petri will outline key features of Firetail and FireSOM and of course we’ll have the possibility to address requirements and questions.
Register for free: https://www.wildlabs.net/event/wildlabs-variety-hour-august
We are happy to present a new part of our tutorial series on Firetail techniques!
In this part you will learn how to
The tutorial has two parts. Firstly, a short lecture to introduce the underlying concepts and secondly, a live demo using Firetail 8 for an illustrated application in practice.
22.1 – Advanced annotation with FireSOM in Firetail 8
ML-assisted prediction of behavioral categories: the revised FireSOM algorithm attends to many details and simplifies daily challenges.
We proudly present Firetail 8!
Firetail features state-of-the art visualization of GPS records, acceleration data, inert magnetic unit (IMU) data and sensor data for wild animal tracking.
Firetail 8 ships with an improved version of FireSOM for machine-learning assisted annotation, featuring new powerful workflows for behavioral prediction and prediction refinement.
refinement via prediction of selected regions, read more details in our revised manual
Firetail 8 can work with the full stack of e-obs sensors and via handling de-facto Movebank standards Technosmart and a wide range of tag manufacturers can be loaded. Firetail 8 now features an advanced detection of bursts and revised import filters for continuously recorded events by Ornitela and Vectronic. All with large-scale, long-term studies in mind.
For your daily field work, Firetail 8 allows you to download map data in advance and process raw tag data whenever there is no network coverage.
interactively inspect goodness-of-fit for IMU orientation
For Movebank data, you can download specific time periods or retrieve data updates to your project.
Firetail 8 is perfectly suited for high-res orientation data. You can now interactively calibrate IMU segments and preview the model’s goodness-of-fit. Select specific time windows and see how your data matches the IMU calibration ellipsoid.
Download Firetail 8 today and be amazed to see your data come to life!
We are glad to present a new part of our tutorial series on Firetail. This time we talk about “FireSOM: AI assisted annotation”
The most recent Firetail Release 7 includes a AI-based segmentation of acceleration data that allows for the annotation of complete datasets in a few steps. It is our mission to offer a simple and intuitive solution. Especially when the challenges are difficult. The new AI feature has sparked impressive feedback in the community of researchers and species conservationists. Questions about how and why arose. Complex issues are best conveyed with videos. Therefore, the decision to make a tutorial was a no-brainer.
Working with annotations is one of the core features of Firetail. The reason is simple: Annotation of data from tagged animals is essential for conservation and for behavioral analysis. This is particularly challenging given huge amounts of data and long-term experiments. Manual annotation is time-consuming and expensive and near impossible to implement in practice.
In this video we showcase “FireSOM: AI assisted annotation”. A machine learning tool assisting researchers to rapidly annotate massive datasets.
Don’t be scared by machine learning terms like feature selection and self-organizing maps. Embrace them!