New App: Seek by iNaturalist

Hi folks,

The iNaturalist team recently got a great opportunity to work with the folks at the Howard Hughes Medical Institute and Tangled Bank Studios on a new app to be released in tandem with their film Backyard Wilderness. They wanted a kid-friendly app that was all about discovering the nature around you. Sounds a lot like iNat, huh? 😀

So we built an iOS app called Seek on top of the iNaturalist APIs, with a few important differences compared to the existing iNaturalist apps. Observations contain some sensitive information that we don't want to reveal about children, including where they are and when they are active (in the US it's actually illegal under the Children’s Online Privacy Protection Act (COPPA) for online platforms to record this information without explicit parental consent). Because the app is designed to be kid-friendly, we don't record any observation information. No activity in this new app becomes a record on, but it uses the same computer vision model as iNaturalist to suggest identifications based on photographs taken by the users. It also suggests species that have been found in the area (based on an obscured location) and recorded on iNaturalist. Seek works because of the observations submitted and identified by the iNaturalist community, so it might work best in areas with active iNaturalist communities.

We're hopeful this app will be fun, and not just for kids. Please try it yourself, and please encourage any of your friends and family who are into nature to give it a try.

It's currently iOS only. You can download it from the iOS App Store here: Seek by iNaturalist. We'll be exploring how to make an Android version soon.


Publicado el marzo 12, 2018 06:55 TARDE por alexshepard alexshepard


Hi there,

I'm already using the app, and I loved it. But I found some wrong Identifications (I'm a Brazilian mastozoologist). For instance, I uploaded a photo of six-banded armadillo (Euphractus sexcinctus) and the application identified as being nine-banded-armadillo (Dasypus novemcinctus). Is there any way we can imrove the application by helping with proper identification?


Publicado por carolinaesteves2 hace casi 6 años

Hi Carol,

Happy you enjoyed using Seek! Unfortunately the machine learning that powers Seek's identifications (and iNaturalist's suggestions) is sometimes wrong, as you found.

Our training dataset is the observations and identifications that have been created by the iNaturalist community. The more correctly identified observations of a species on iNaturalist, the more accurate the computer vision will be. So I would encourage you to get active in the iNaturalist community: make and upload observations using the main iNaturalist app or using our website, identify other people's observations, join our Google Group to share your expertise as a professional taxonomist, etc. In addition to improving the quality of iNaturalist with your very presence, every interaction on the iNaturalist site increases the quality of the machine learning system a little bit.

Thank you,

Publicado por alexshepard hace casi 6 años

Hi Alex,

Now I understand how the application works, very interesting. I will participate more actively on iNaturalist community with my contributions.

Thank you very much for the information.

Publicado por carolinaesteves2 hace casi 6 años

Hi Alex ! You told us 9 months ago that a android version was coming ! Everybody is waiting ! When will it be official ? Thank you

Publicado por jonathan148 hace más de 5 años

Hi @jonathan148, we hired a new developer who's taken over work on Seek, and she's working on getting Seek working on Android. You can read more here:

Publicado por alexshepard hace más de 5 años

I’m impressed by the way this app identified a number of tropical taxa (getting at least to family level). However, there are a few things I would like to suggest:

provide a way to link from the identification to an appropriate website
allow users to choose scientific names by preference
disable game challenges etc for users who want to use this as an identification tool only
provide a way to quickly refind photos in the camera roll that have already been identified
provide a feedback mechanism for users to correct identifications etc
allow the spp to make use of EXIF GPS data to assist in identification
That’ll do for starters - thanks!

Publicado por ronaldorenstein hace casi 5 años

@ronaldorenstein Thanks for the feedback! Since this post was published last year, a whole new version of Seek has been developed and released. I'm no longer the developer of Seek, and so my blog post isn't the best place to provide feedback. Please post your comments to for further discussion. I'd also encourage you to also consider that for many of your suggestions, the main iNaturalist app is a better fit.

Publicado por alexshepard hace casi 5 años

impress app really very complete, I love it

Publicado por teresasans hace casi 4 años

Hi @alexshepard and @albullington ,

I am interested in using the CNN behind seek for a project of ours related to research in plant-pollinator interactions. We want to take thousands of photos with a camera system installed in the field, focused on flowers and capture images of insects. Would it be possible to download and run your CNN or give us some rights to use it for research purposes?
Another related question - is the code used for training the CNN public on GitHub or somewhere online?


Publicado por valentin_stefan hace casi 3 años

Hi @valentin_stefan, can you please email with more details?

Publicado por alexshepard hace casi 3 años

Agregar un comentario

Acceder o Crear una cuenta para agregar comentarios.