We have developed a novel, state-of-the-art algorithm for detecting saccades and microsaccades in eye movement traces.
The algorithm, based on deep neural networks, achieves human-level performance, and it is also able to detect post-saccadic oscillations as well as eye blinks.
The algorithm outperforms existing algorithms on benchmark data sets, as well as on our in-house data sets.
We have made all code for the algorithm available for anyone to use, and we will also provide our in-house data sets soon. In the near future, we will also create a web-service, such that anyone can upload data and have the deep neural network detect eye movement events.
You can read about the algorithm here.
And, you can download the code here.