[image] What Image(s) Are Used for "Training" Computer Vision? our computer vision model is not trained to recognize a certain taxon - it is trained to recognize iNaturalist photos of a certain taxon @tiwane - can this training and observation entry be modified somewhat? A real challenge for…

It’s a bit of a vicious circle – to be able to recognize and annotate life stages correctly, the AI needs a suitable quantity of annotated material. So it seems to me that there is no way to avoid the necessity of humans annotating observations to start with. Another issue is that it is currently n…

[image] spiphany: Another issue is that it is currently not possible to correct annotations unless you are the observer or annotator. I was going to mention that as well. Technically, we currently have a trial for that concept within iNat itself. Many, Many, users just use the suggested ID,…

Theoretically a great idea. However, we’d need to change how we correct annotations (as noted above) and getting a good base of specimens for training would be a challenge. Might have to be implemented taxon by taxon. I’m ambivalent about actually implementing this. (Fortunately, I won’t be invo…

I can see this working well, if implemented on a very small, hand-selected group of taxa, for example: life stages in Lepidoptera (this was mentioned earlier) sexes in strongly sexually dimorphic species, such as mallards or northern cardinals. I’m not sure how it would work for more gradual thin…

[image] someplant: a very small, hand-selected group of taxa This would be the only way I’d vote for this- I’d want to see it rolled out on a small group of taxa first. My main concern is that which annotations are relevant/correct depends upon the identification of the observation. So if th…

I came here to find out if anyone had ever suggested this. I would think it’s an obvious and hugely useful strategy. I know it would need to have large numbers of annotated observations to train on, but by now I’d have thought we have that for many species. I’ve spent many hours annotating the life…

[image] paul_dennehy: If the CV is given that the photo shows a monarch, then annotating it as adult or larva seems like really easy photo recognition. If the CV isn’t given what the identity is as a starting point, then CV annotations might be a huge mess. I wouldn’t necessarily see this as …

Agreed, [image] paul_dennehy: you’d have to have the ID suggestion box include the annotation in what you click on when you take the CV identification (ie top suggestions says “Monarch larva” to click on, rather than just “Monarch”)

I think first, we need to educate observers (and identifiers) on the importance of annotations.

I think a better (separate?) interface could also speed up things significantly - like “gamifying” it with arrow left/right and having the next observation showing up instantly. I have been playing with that idea for detecting bird bands; like making a small webapp to quickly assess all observations…

Use computer vision to annotate observations

Feature Requests

ajarmund January 12, 2026, 7:43pm 66

Nice, thanks, I didn’t know!

Topic		Replies	Views
Adding annotations - does anyone use them? General	91	2182	March 25, 2026
Annotations: Male/female, etc.—How to get more action? General question	75	4106	December 19, 2020
Annotation Sunday General	57	2048	March 24, 2025
Lack of Annotations? General	29	1759	September 25, 2023
Add annotations when creating observations Feature Requests web , mobile-app	26	5804	November 11, 2025

Use computer vision to annotate observations

Related topics