Adding captive organisms to help the algorithm?

Is there any value to adding captive organism observations to help train the algorithm? I have pet giant hissing cockroaches and for a standard adult, the iNat algorithm knows exactly what they are. (I haven’t actually uploaded any observations, but I like starting observations of captive things or random objects just to see what the ID algorithm thinks they are ;-) ) But one recently molted, and I got a great shot of it looking all white and ghostly. The algorithm was utterly confused. Would it be useful to add that as an actual observation to help people who might encounter one in the wild?

4 Likes

I think captive observations definitely have value for other iNat users.

As for whether they help with CV, well, that’s a bit more iffy, because it is “trained” on a random selection of images, and you have no way of knowing whether yours will be one of the selected images or not:
https://forum.inaturalist.org/t/what-image-s-are-used-for-training-computer-vision/3307

1 Like

It may or may not help the CV, but I think it is kind of cool to add observations that document interesting facets of nature - whether or not they get to research grade.

9 Likes

Thank you!