New CV image matching selects photos from inactive taxa

Bug Reports
1

Please fill out the following sections to the best of your ability, it will help us investigate bugs if we have this information at the outset. Screenshots are especially helpful, so please provide those if you can.

Platform (Android, iOS, Website): Website

App version number, if a mobile app issue (shown under Settings or About):

Browser, if a website issue (Firefox, Chrome, etc) : Firefox

URLs (aka web addresses) of any relevant observations or pages: https://www.inaturalist.org/observations/165144329

Screenshots of what you are seeing (instructions for taking a screenshot on computers and mobile devices: https://www.take-a-screenshot.org/):

image
image1920Ă—1080 302 KB

Description of problem (please provide a set of steps we can use to replicate the issue, and make as many as you need.):

Step 1: Go to the observation of the species Americina adusta linked above.

Step 2: Get CV suggestions. Since Americina is not in the CV, the similar-looking genus Cordilura is suggested as “pretty sure”.

Step 3: The photo that comes up is nevertheless a picture of Americina adusta! I assume this is coming from the inactive taxon Cordilura adusta, which was left behind after removing the species from this genus on iNat.

IMO, photos associated with inactive taxa should not be eligible for CV image matching, because it can be misleading when taxon changes have occurred.

2 Likes
2

You could remove the image from the inactive taxon?
Especially since the exact same image is being used for the active taxon. One of them must be wrong.

3

I don’t think inactive taxa are removed from the CV until the next update?

4

I just removed the one available image from taxon page for the inactive Cordilura adusta. On the cited observation, the CV still suggests that ID, but it may take time for that photo removal to cycle through–maybe for the next CV training.

1 Like
5

I have had the same problem when a subspecies was removed and a new species was added, but both had the same pictures. My temporary solution was removing the picture from the inactive taxon, but I think the bigger issue is the similar photo update drawing from subspecies photos.

2 Likes
6

This is a 9-month-old taxon change, and one for a species that’s never been in the CV. I don’t think any kind of lag is responsible.

I’m sure there are lots of inactive taxa inside genera (left over from every genus split that’s ever happened), which are hard to find even if you’re looking for them. Up till now, it didn’t matter at all what photos were attached to those old names, but now they can show up in the wrong place and mislead observers trying to use the CV thoughtfully.

Yes, removing the photo from “Cordilura adusta” should fix the problem in this instance, but I don’t think it’s reasonable for people to have to hunt down every inactive taxon and remove its photos. There’s no reason for inactive taxa to be included in this feature at all.

5 Likes
7

Does this point to the need to make a Feature Request that Inactive Taxa be excluded from consideration or outputs by CV? I’m not a programmer, so I don’t know how complex that action might be.

2 Likes
8

here’s another just-flagged example of what sounds like the same issue (photos of inactive taxa, in this case Calonectris diomedea borealis, being brought up as similar photos); i’ve encountered the same thing recently with a photo of the inactive Sula leucogaster brewsteri being displayed for Sula leucogaster, though i unfortunately neglected to save the observation in question (if i refind it i’ll add the link)

1 Like
9

another instance here https://www.inaturalist.org/observations/148265641 - the photo showing up for Campylorhynchus rufinucha is the same photo showing up for the recently-split C. capistratus, from a now-inactive subspecies of the former

1 Like