Hello everyone,
For my PhD, I’m working on a accelerometer decoder using flux. But first I need to annotate videos of animal behaviour, and I’d like to have some software or julia code that allows me te predefine the classes of behaviours to annotate, assign them to numbers on my keyboard, and then just press those buttons as I watch the video to annotate it. And ideally that would yield a csv with the times of when each behaviour happened. There’s nothing on google that I could find that fits this description.
There are a few free and/or open source labelling solutions out there. https://labelstud.io/ and https://diffgram.com I know off the top of my head, but I’m sure there are many more.
If you need to annotate a lot of videos like that, then it might be convenient / less error prone to be able to do that by saying the the class instead of typing the right number? If so, you could use JustSayIt to define a mapping of your classes to functions or keyboard shortcuts which do whatever you need to do: https://github.com/omlins/JustSayIt.jl
Concretely see here how trivial it is to do such a mapping and start JustSayIt using it: Usage · JustSayIt.jl
The functions or keyboard shortcuts mapped can then do whatever you want, i.e. write direclty into a csv or trigger some actions in the softwares mentioned by @ToucheSir …