XnView Software

I think the tagging system is outdated and should get a major overhaul. I believe it would be much easer to use if:

1. No more right click and trying to find the correct category from the context menu
2. The category icon which can be displayed on top of a thumbnail should be clickable and open the tagging system
3. When opened, there should be option select one or even multiple categories by mouse but way more efficient would be an additional system to simply enter the names of the categories this file should be associated with. For example: If I have a picture file showing my wife, my kids in year 2010 during vacation in Italy, I would like to have a categories field where I can simply type: "wife, kids, 2010, vacation, Italy" and the tagging system should recognise the comma seperated words as tags. If the tags do not exist yet, then the system should create them.

This way it would be much easier to tag a large number of files with multiple tags.

Also it would be great to have some kind of auto tagging feature. The system could suggest auto tags from the filename and the year the file has been created.

I separately came to think of the same idea. As a result, I also suggest an 'auto-tagging' feature, albeit in a different name, in a freshly made thread. Though I fleshed out the idea a bit more by adding delimiters, a whitelist/blacklist, etc...

I can only hope to see this kind of thing implemented soon, but alas, it will probably take a long time.

IDImager does something more advanced/better than your solution. However, its successor (PhotoSupreme) as well as almost all other programs (Lr, ACDSee, Zoner etc.) have a separate panel like XnView MP.

And more XnView has two: Categories (you can toggle the focus on it with F9 - did you knew that? - have the Categories panel on display and press F9 to see how the focus switch form Thumbs pane to it and back) and Category Sets.

MrSwordFish wrote:I separately came to think of the same idea. As a result, I also suggest an 'auto-tagging' feature, albeit in a different name, in a freshly made thread. Though I fleshed out the idea a bit more by adding delimiters, a whitelist/blacklist, etc...

I can only hope to see this kind of thing implemented soon, but alas, it will probably take a long time.

Which thread is that?

jadO wrote:Which thread is that?

Importing function for mass categorization

jadO wrote:I think the tagging system is outdated and should get a major overhaul.

Agree. Please see my suggestion for making tagging much simpler for the vast majority of users.

jadO wrote:3. When opened, there should be option select one or even multiple categories by mouse but way more efficient would be an additional system to simply enter the names of the categories this file should be associated with. For example: If I have a picture file showing my wife, my kids in year 2010 during vacation in Italy, I would like to have a categories field where I can simply type: "wife, kids, 2010, vacation, Italy" and the tagging system should recognise [sic] the comma seperated [sic] words as tags. If the tags do not exist yet, then the system should create them. This way it would be much easier to tag a large number of files with multiple tags.

Yes. This is how gThumb and F-Spot do tag autocompletion, and it's much faster to tag many images with those apps than with XnView MP, using the keyboard.

jadO wrote:Also it would be great to have some kind of auto tagging feature. The system could suggest auto tags from the filename and the year the file has been created.

A nice option would be to call Google's free Vision API and automatically detect objects in the image. The API is very, very simple to use - just one HTTP call to https://vision.googleapis.com/v1/images:annotate, and you get a JSON output with the entities detected in the image.

Yes. This is how gThumb and F-Spot do tag autocompletion, and it's much faster to tag many images with those apps than with XnView MP, using the keyboard.

Already there. Use the Categories Pane (F9 to switch focus from/to it) and write down „vacantion”[enter]„kids”[enter]„2017”[enter]

In this way you type faster (Enter is easier to press than „;”) and you can have already incremental search, Auto-Correct (for ex. „kids” for „children”) etc.

HTH

A nice option would be to call Google's free Vision API and automatically detect objects in the image. The API is very, very simple to use - just one HTTP call to https://vision.googleapis.com/v1/images:annotate, and you get a JSON output with the entities detected in the image.

Good idea. However it needs synchronous inet link. We need thinking to make it more flexible - for example async batch tagging.

m.Th. wrote:

Yes. This is how gThumb and F-Spot do tag autocompletion, and it's much faster to tag many images with those apps than with XnView MP, using the keyboard.

Already there. Use the Categories Pane (F9 to switch focus from/to it) and write down „vacantion”[enter]„kids”[enter]„2017”[enter]

F9 doesn't work in preview or full-screen mode. If my workflow is to view image after image in full-screen mode and tag it, what option do I have?

dandv wrote:

m.Th. wrote:

Yes. This is how gThumb and F-Spot do tag autocompletion, and it's much faster to tag many images with those apps than with XnView MP, using the keyboard.

Already there. Use the Categories Pane (F9 to switch focus from/to it) and write down „vacantion”[enter]„kids”[enter]„2017”[enter]

F9 doesn't work in preview or full-screen mode. If my workflow is to view image after image in full-screen mode and tag it, what option do I have?

Hmmm... THE best option by far is to use the dual-screen mode. It will make your life waaay easier in many aspects (not only WRT XnView MP) - graphics, programming, inet/email etc. If you have a two monitor setup (a dual-head VGA powering two monitors) you can configure XnView MP to use the 2nd monitor for the Full-Screen mode and then F9 will work flawlessly.