# Species Photos Required to Train Vision **URL:** \ **Category:** General\ **Created:** [October 22, 2021, 9:38pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379 "2021-10-22T21:38:37Z")\ **Posts on this page:** 16\ **Page:** 1
**Author:** ![SteveMcBill](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/stevemcbill/32/50876_2.png) [@SteveMcBill](https://forum.inaturalist.org/u/SteveMcBill)\ **Post date:** [October 22, 2021, 9:38pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/1 "2021-10-22T21:38:37Z")
How would one go about finding which species need observing/recording in order to provide photos for future training of Computer Vision ?? Many thanks for any insights Cheers, Steve ---
**Author:** ![marina\_gorbunova](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/marina_gorbunova/32/31454_2.png) [@marina\_gorbunova](https://forum.inaturalist.org/u/marina_gorbunova)\ **Post date:** [October 22, 2021, 9:47pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/2 "2021-10-22T21:47:17Z")
On species page there’s a badge if it’s included in cv: ![image](https://us1.discourse-cdn.com/flex016/uploads/inaturalist/original/3X/b/2/b2e9f0f951efffda8f0a24dcd92d827e97a7236b.png) So if there’s none this species probably needs more photos, anyway, new photos are always welcome! ---
**Author:** ![susanhewitt](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/susanhewitt/32/887_2.png) [@susanhewitt](https://forum.inaturalist.org/u/susanhewitt)\ **Post date:** [October 22, 2021, 10:23pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/3 "2021-10-22T22:23:43Z")
Also we need a lot of good identification work, because there may already be photos of a species, but if they are not identified, then the images can’t yet be fed to the CV. ---
**Author:** ![trscavo](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/trscavo/32/3386_2.png) [@trscavo](https://forum.inaturalist.org/u/trscavo)\ **Post date:** [October 22, 2021, 11:41pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/4 "2021-10-22T23:41:23Z")
> [@susanhewitt](#): > > there may already be photos of a species, but if they are not identified, then the images can’t yet be fed to the CV. I don’t think that’s true. An observation need not be Research Grade to be included in the training. An observation doesn’t need a community ID either (one ID is sufficient). It doesn’t even need to be verifiable (e.g., not wild observations are eligible). ---
**Author:** ![marina\_gorbunova](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/marina_gorbunova/32/31454_2.png) [@marina\_gorbunova](https://forum.inaturalist.org/u/marina_gorbunova)\ **Post date:** [October 22, 2021, 11:50pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/5 "2021-10-22T23:50:39Z")
If it’s not ided as correct taxon or left at family level or higher it’s not useful for cv, and that’s the case for tons of taxa and thousands of observations. ---
**Author:** ![sbushes](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/sbushes/32/8217_2.png) [@sbushes](https://forum.inaturalist.org/u/sbushes)\ **Post date:** [October 23, 2021, 12:55am UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/6 "2021-10-23T00:55:23Z")
Aside from the badge @marina_gorbunova mentions, there is also a feature request to make it a bit easier to get a broader overview in future : [https://forum.inaturalist.org/t/add-label-on-drop-down-menu-for-species-in-computer-vision-training-model-to-give-quick-overview/24874](https://forum.inaturalist.org/t/add-label-on-drop-down-menu-for-species-in-computer-vision-training-model-to-give-quick-overview/24874) ---
**Author:** ![pisum](https://avatars.discourse-cdn.com/v4/letter/p/91b2a8/32.png) [@pisum](https://forum.inaturalist.org/u/pisum)\ **Post date:** [October 23, 2021, 3:20am UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/7 "2021-10-23T03:20:05Z")
right now the /v1/taxa/{id} API endpoint returns a “vision” field in its results, but the v1/taxa endpoint does not. if the latter did also return that field, then it would be relatively easy to get/view that information for a bunch of taxa. right now, you’d have to either go in with a list of taxon IDs, or you’d have to query to get a list of IDs and then query by that list of IDs. it’s possible to do this, but i’m not motivated enough myself to do it. ---
**Author:** ![sbushes](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/sbushes/32/8217_2.png) [@sbushes](https://forum.inaturalist.org/u/sbushes)\ **Post date:** [October 23, 2021, 10:33am UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/9 "2021-10-23T10:33:15Z")
@pisum Nice thanks for the tip, maybe I will take a look myself @brian\_d The number of photos/observations species require to be included is on the help page [here](https://www.inaturalist.org/pages/help#cv-taxa) > “as of the model released in March 2020, taxa included in the computer vision training set must have at least 100 observations, at least 50 of which must have a community ID. Photos for training are randomly selected from among the qualifying iNaturalist observations (that is, it is not only the first image of an observation that may be used for training)” ---
**Author:** ![mrtnlowr](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/mrtnlowr/32/6890_2.png) [@mrtnlowr](https://forum.inaturalist.org/u/mrtnlowr)\ **Post date:** [October 23, 2021, 10:50am UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/10 "2021-10-23T10:50:06Z")
It wouldn’t be too difficult to modify [this code](https://jumear.github.io/stirfry/iNat_taxa_conservation_status.html) I wrote for querying conservation status to also provide the taxon CV flag. It uses the algorithm outlined here: > [@pisum](#): > > … query to get a list of IDs and then query by that list of IDs Cheers! ---
**Author:** ![susanhewitt](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/susanhewitt/32/887_2.png) [@susanhewitt](https://forum.inaturalist.org/u/susanhewitt)\ **Post date:** [October 23, 2021, 4:54pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/11 "2021-10-23T16:54:40Z")
I did not say the photos of a species had to be Research Grade, just that they had to be identified. ---
**Author:** ![matthias22](https://avatars.discourse-cdn.com/v4/letter/m/aeb1de/32.png) [@matthias22](https://forum.inaturalist.org/u/matthias22)\ **Post date:** [October 23, 2021, 5:18pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/12 "2021-10-23T17:18:36Z")
Would it be possible for experts to review the photo set that is used for training in order to weed out misidentifications? I think that could speed up the process significantly. I am talking about groups that are hard to ID and still largely inaccessible because of lack of expertise in the community (e.g., I work on wasps). ---
**Author:** ![matthias22](https://avatars.discourse-cdn.com/v4/letter/m/aeb1de/32.png) [@matthias22](https://forum.inaturalist.org/u/matthias22)\ **Post date:** [October 23, 2021, 5:42pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/14 "2021-10-23T17:42:25Z")
Another interesting question is how life stages and sexes are being handled. In some cases it is pretty straightforward such as caterpillars and butterflies. I assume two separate sets of images are being used to train CV for each of the two. It gets more complicated with the sexes or with species that gradually transform from juveniles into adults. Depending on the species, sexual dimorphism can vary from extreme to imperceptible, with every possible intermediate condition. I assume _a priori_ decisions have to be made on how many separate entities CV will be trained for. For species with extreme sexual dimorphism it is obvious that there have to be separate sets of images to train CV. But where is the cut-off for species with moderate sexual dimorphism? Many insects fall into this category. To the untrained eye males and females may look quite similar yet taxonomists often use different diagnostic characters for each sex to get to a species ID. ---
**Author:** ![fluffyinca](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/fluffyinca/32/10786_2.png) [@fluffyinca](https://forum.inaturalist.org/u/fluffyinca)\ **Post date:** [October 23, 2021, 7:55pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/15 "2021-10-23T19:55:08Z")
> [@matthias22](#): > > I assume two separate sets of images are being used to train CV for each of the two I was under the impression that they are not done separately. From what I’ve read on various forum threads, it sounds like the training model randomly selects photos from among the eligible observations of that taxon, regardless of life stage or sex. But I could be wrong. Doing them separately would make sense in cases like the ones you described. ---
**Author:** ![susanhewitt](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/susanhewitt/32/887_2.png) [@susanhewitt](https://forum.inaturalist.org/u/susanhewitt)\ **Post date:** [October 24, 2021, 1:34pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/17 "2021-10-24T13:34:22Z")
The life stages and sexes are not fed separately to the CV. It does not learn in the same way that a human learns, so variations like that do not confuse it. ---
**Author:** ![sbushes](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.inaturalist.org/sbushes/32/8217_2.png) [@sbushes](https://forum.inaturalist.org/u/sbushes)\ **Post date:** [October 24, 2021, 4:59pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/18 "2021-10-24T16:59:16Z")
Realised it’s pretty straightforward to look and see which species are included just using the observations portal of course. We can see we need to find some more Proteroiulus fuscus to gear up in European Blanuilidae for example! Only Blaniulus guttulatus has \> 50RG and \> 100 total obs in this family in Europe at present, likely leading to oversuggestion of Blaniulus sp. * * * RESEARCH GRADE ![Screenshot 2021-10-24 at 17.53.50](https://us1.discourse-cdn.com/flex016/uploads/inaturalist/original/3X/9/1/91a96ff310d87b88b9533c92109da169446bbe59.jpeg) * * * ALL ![Screenshot 2021-10-24 at 17.53.29](https://us1.discourse-cdn.com/flex016/uploads/inaturalist/original/3X/3/d/3d5d0cd3f9835af8cba92de45c28dad217189252.jpeg) ---
**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/inaturalist/original/2X/7/7831452cf7f0d653ff1ffd34176334b7ef1b6a17.png) [@system](https://forum.inaturalist.org/u/system)\ **Post date:** [December 23, 2021, 5:00pm UTC](https://forum.inaturalist.org/t/species-photos-required-to-train-vision/27379/19 "2021-12-23T17:00:14Z")
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.