Does the iNat computer vision use all photos in an observation for suggestions or just the first?

I couldn’t find this in the forums. I don’t think the CV uses all photos I upload because I just uploaded some mushrooms with multiple photos and ran a test.

I took a few close photos of some mushrooms with my Panasonic FZ80 camera. I then took a photo with my phone to get the GPS coordinates.

I uploaded the phone photo first so I the location would be in the observation, and the CV gave me…something not accurate. That’s fine because it was a distant photo. I then uploaded all the additional photos from my camera to that saved observation and it still couldn’t give me a good suggestion.

I did get a good suggestion when I changed the order of the photos where one from my Panasonic camera was first. So, I’m thinking the CV only uses the first photo for a suggestion.

It has been discussed in the forum before, the CV only uses the first photo.

https://forum.inaturalist.org/t/use-computer-vision-on-each-photo-in-an-observation/4210

3 Likes

Note in the new app you can tap on a photo and get suggestions for it.

4 Likes