[digiKam-users] a wild idea or wish
Locked
[digiKam-users] a wild idea or wish
 
|
I have no idea if this is possible, if the necessary AI exits
opensource, but as there is face-recognition in digikam, I dare to throw in my idea: google, facebook and others can recognize the contents of images. If it's a house, a portrait, a beach, a tree, if male/female, if she has nipples or if cheeks are visible (no joke, both facebook and google can and do - with lots of errors, though), etc. Would it be possible that digikam can analyze and categorize images and make them searchable by content? So that they are searchable by content. That would be fantastic. Imagine! -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer https://www.daniel-bauer.com |
|
|
I would like to second this.
Some sort of "object detection" would be a great addition to DigiKam. For example, "OwnPhoto" claims to have it already implemented "Detect objects in photos, and make them searchable by objects" using Object detection: densecap, places365. See https://github.com/hooram/ownphotos However the development of OwnPhoto seems somehow stuck in 2019 and it might exist a fork already. "Photoprism" seems an alternative using some AI stuff for image classification, see: https://docs.photoprism.org/developer-guide/metadata/classification/ and in more general https://github.com/photoprism/photoprism So, somehow object detection seems to be implemented in some open source products. On 7/27/20 10:48 PM, Daniel Bauer wrote: > I have no idea if this is possible, if the necessary AI exits > opensource, but as there is face-recognition in digikam, I dare to throw > in my idea: > > google, facebook and others can recognize the contents of images. If > it's a house, a portrait, a beach, a tree, if male/female, if she has > nipples or if cheeks are visible (no joke, both facebook and google can > and do - with lots of errors, though), etc. > > Would it be possible that digikam can analyze and categorize images and > make them searchable by content? So that they are searchable by content. > > That would be fantastic. Imagine! |
|
|
Hi all,
This kind of photo contents analysis exist already with this CLI tool : https://github.com/scheckmedia/photils-cli Photils API is very simple to handle and can be used in a dedicated digiKam plugin. The image content is analyzed locally, a fingerprint is computed locally by a AI engine, the photo still on computer. AI parse the fingerprint and return a list of keywords. Et voilà, you have the tags to validate in client application. We plan to include photils as extra digiKam plugin later 7.2.0. Please be patient... Best Gilles Caulier Le mar. 28 juil. 2020 à 09:15, <[hidden email]> a écrit : > > I would like to second this. > Some sort of "object detection" would be a great addition to DigiKam. > For example, "OwnPhoto" claims to have it already implemented "Detect > objects in photos, and make them searchable by objects" using Object > detection: densecap, places365. See https://github.com/hooram/ownphotos > However the development of OwnPhoto seems somehow stuck in 2019 and it > might exist a fork already. > "Photoprism" seems an alternative using some AI stuff for image > classification, see: > https://docs.photoprism.org/developer-guide/metadata/classification/ and > in more general https://github.com/photoprism/photoprism > So, somehow object detection seems to be implemented in some open source > products. > > > On 7/27/20 10:48 PM, Daniel Bauer wrote: > > I have no idea if this is possible, if the necessary AI exits > > opensource, but as there is face-recognition in digikam, I dare to throw > > in my idea: > > > > google, facebook and others can recognize the contents of images. If > > it's a house, a portrait, a beach, a tree, if male/female, if she has > > nipples or if cheeks are visible (no joke, both facebook and google can > > and do - with lots of errors, though), etc. > > > > Would it be possible that digikam can analyze and categorize images and > > make them searchable by content? So that they are searchable by content. > > > > That would be fantastic. Imagine! > |
|
|
I would totally love to see more AI tools included in Digikam. Like, being able to search for objects or the content of a Picture without having to manually tag them. E.g. searching photos of beaches, or pictures containing dogs but not cats. Well, as long as the processing is carried offline (or with my permission). I don’t want my personal pictures uploded on some server without my knowledge... De: [hidden email] Hi all, |
|
|
In reply to this post by Gilles Caulier-4
On 28/07/2020 11:04, Gilles Caulier
wrote:
We plan to include photils as extra digiKam plugin later 7.2.0. Please be patient... That sounds great. I tried syncing my photos with Google just for this feature. But it would be much better having it in digiKam. Yours patiently Dougie |
|
|
In reply to this post by woenx
Den tir. 28. jul. 2020 kl. 20.38 skrev Marc Palaus <[hidden email]>:
Absolutely! All processing should happen locally. Anything else is unacceptable. BR Thomas |
|
|
In reply to this post by Gilles Caulier-4
Am 28.07.20 um 12:04 schrieb Gilles Caulier: > Hi all, > > This kind of photo contents analysis exist already with this CLI tool : > > https://github.com/scheckmedia/photils-cli > > Photils API is very simple to handle and can be used in a dedicated > digiKam plugin. The image content is analyzed locally, a fingerprint > is computed locally by a AI engine, the photo still on computer. AI > parse the fingerprint and return a list of keywords. Et voilà, you > have the tags to validate in client application. > > We plan to include photils as extra digiKam plugin later 7.2.0. Please > be patient... Sounds good :-) As I am not very patient I downloaded photils but unfortunately it needs GLIBC 2.27 where I (on opensuse 15.1) have 2.26. So I can't try it... -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer https://www.daniel-bauer.com |
| Free forum by Nabble | Edit this page |