Newbie questions that I didn't manage to figure out myself

[plastikman]

Hi everyone. I have been looking for a proper DAM for a while. I am currently testing IMatch 2020 and I really like it so far. One of the best things about it must be the extensive contextual help system that can be triggered with F1 and the general application look, feel and feature set.

However, I still have some questions:

• Is it possible to see zoomed in faces in the people tab? I find it very hard if not difficult to confirm large numbers of images for a person if I cannot see which person was tagged with the name if multiple people are on a photo
• In which scenario does it make sense to rotate the face recognition region? In many cases I find that rotation is unnecessary but sometimes it helps but this is really random and unpredictable
• Microsoft Azure free tier doesn't work for me. It keeps throwing an error (which I will share in the bug section)
• I have been tagging my images with Google Vision. When doing batches, it often gets stuck. I believe that this happens when there is an image in the batch for which Google Vision finds/has no keywords. This means I had to auto tag in many very small batches making it very inefficient. The Any Vision lightroom plugin also works with Google Vision and has no problem whatsoever tagging 10k images in one batch
• It would be great if the auto tagger keywords can be stored in another field than keywords since I like them for aiding my searches but not in the main keywords field since they are often wrong and they make it harder to see my own keywords (see also Any Vision for how they did this)
• Any Vision can also rebuild the keywords from the custom plug-in metadata fields without re-running the queries to Google Vision servers. This is a great feature that gives a lot of flexibility especially when you change certain settings for the batch processing.
• With regards to categories: I want to make a data-driven category for all the keywords that start with Google Vision|XXXX. Then I want to have the grouping based on the first letter. How would I go about this?
• The same for all people which I would like to group under their family name. How do I filter out a sub-section of the Thesaurus (the group is "people" under which I have "Family A", "Family B" under which the individuals are listed
• I have another keyword group in my Thesaurus called "Events" under which I have "Wedding", "Daytrip", "Holiday" etc. I want to make a category which lists these types of events. How would I setup that filter, I can't work it out (it is similar to the examples above e.g., finding a hierarchical keyword and filtering everything under it while hiding the first part
• For my .NEF files I get a large and detailed preview. However, for my .DNG files the preview is much smaller and also way less sharper which makes it hard to judge/rate photos quickly. Is there a way to create bigger previews for my DNG files?
• I am also testing a competing product. It has a very handy feature where I can select a group of images and "batch open" them in my editor of choice, "DxO PhotoLab" which automatically creates a "project" with the selected images that can be from many different folders on my harddisk. Is there a way to set this up? Currently it only shows "Open in FastRawViewer" and I cannot find a way to add other applications. Also, when multiple files are selected the "Open with..." command under the dropdown is greyed out.
• Is there a way to see at glance if a photo is part of an event in the main folder view?
• Is there a keyboard shortcut to trigger the "Face Annotation" button in the full screen viewer? I find myself clicking it an awful lot which brings down the speed.
• Is there a way to automatically assign keyword X when I manually assign keyword Y? e.g., when I assign my daughter's name also assign "girl"
• Would it be possible to have a keyboard shortcut to trigger auto tagger based on the last used settings?
• Some of my images are showing in the wrong orientation. When I rotate them in full-screen viewer, they still show as unrotated in the grid/folder view. If I rotate them in the grid/folder view, the full screen viewer often hangs on loading the preview images, again mostly with the converted .DNG files and not with my newer .NEF files.
• I want to make a category with images that are not labeled "Green". However, whatever I try with regards to adding NOT at different places in the formula template in IMatch, it doesn't work for me. What am I doing wrong?

Sorry for the barrage of questions but I am seriously evaluating this software before I purchase. So far I am very pleased with what I see. I hope some fellow forum members can help so Mario doesn't have to answer all these questions, which I didn't manage to figure out with the help system.

[jch2103]

I'll tackle a few of your questions:

Is it possible to see zoomed in faces in the people tab? I find it very hard if not difficult to confirm large numbers of images for a person if I cannot see which person was tagged with the name if multiple people are on a photo

Look for Zoom in the Help for the Viewer: http://127.0.0.1:50519/help/index.html#viewer.htm

With regards to categories: I want to make a data-driven category for all the keywords that start with Google Vision|XXXX. Then I want to have the grouping based on the first letter. How would I go about this?

See Help for Data-driven categories; search for 'Start and Length', but read the entire topic for an overview and detailed explanation and examples.

For my .NEF files I get a large and detailed preview. However, for my .DNG files the preview is much smaller and also way less sharper which makes it hard to judge/rate photos quickly. Is there a way to create bigger previews for my DNG files?

Check the Forum; this is an issues that's been raised a number of times because DNG doesn't always embed full-size previews.

I am also testing a competing product. It has a very handy feature where I can select a group of images and "batch open" them in my editor of choice, "DxO PhotoLab" which automatically creates a "project" with the selected images that can be from many different folders on my harddisk. Is there a way to set this up? Currently it only shows "Open in FastRawViewer" and I cannot find a way to add other applications. Also, when multiple files are selected the "Open with..." command under the dropdown is greyed out.

I and a number of others here also use DxO. The answer is easy: Make a selection in IMatch and just drag and drop into DxO. That will create a new Project. There are also a number of ways to aggregate files from multiple folders, including the IMatch Clipboard http://127.0.0.1:50519/help/index.html#fw_basics.htm

Good luck!

[plastikman]

For the data-driven categories, what I am trying to figure out is how to filter levels within keywords.

So if may keyword structure is:

Group keyword A | Item keyword 1, Item keyword 2, Item keyword 3
Group keyword B | Item keyword 1, Item keyword 2, Item keyword 3
Group keyword C | Item keyword 1, Item keyword 2, Item keyword 3

I want to populate level 1 of my data-driven category with the Item keyword, not with the group keywords.

Then I want to sort those Item keywords based on their first letter, give them a distinct color etc. I cannot figure this out from the various help section about categories, data-driven categories and formulas. But then again, I am new to this.

Thanks for your input on DxO. I will test it out today and see how it works for me.

[Mario]

20 questions in one post. This has to be a new record 
Asking so many questions in one thread makes it very hard for other users to help you.
Much better to ask one question per thread.
Else all this will end in a unreadable heap of quotes and partial answers.

However, for my .DNG files the preview is much smaller a

IMatch uses the embedded preview if it large enough. The minimum size is set under Edit > Preferences > Cache. It is by default 1920 pixels.
If the embedded preview does not meet the minimal size, IMatch falls back to developing the full DNG RAW via the installed WIC codec.

To check, run a WIC Diagnostics on one of your DNG files. This shows you how large the embedded preview is, if the DNG format is handled by the installed WIC codec (there are so many different DNG variants in use these days...).

If AutoTagger gets stuck, provide at least a minimum of information. Like the IMatch logfile from that session log file.
Google may also just throttle your account for abuse prevention or because too many requests are coming in a too short time. I'm not aware of any Lr plug-ins, I don't use them.
So far you are the only person having reported this.

I am also testing a competing product. It has a very handy feature where I can select a group of images and "batch open" them in my editor of choice, "DxO PhotoLab"

This has been asked and answered so many times, I can only refer you to the community search engine. Just search or Dxo.
jch also has explained that above again.

See also Opening Files in Other Applications in the IMatch help.
I found that by searching for the term open.

You may also want to look at Favorites

Is there a way to automatically assign keyword X when I manually assign keyword Y? e.g., when I assign my daughter's name also assign "girl"

Synonyms in the Thesaurus

rotate them in full-screen viewer, they still show as unrotated in the grid/folder view. If I rotate them in the grid/folder view, the full screen viewer often hangs on loading the preview images, again mostly with the converted

This usually indicates a damaged file. When the Viewer hangs on a rotated image, the problem is inside the WIC codec (Microsoft) or the DirectX subsystem IMatch uses. This is a known bug but Microsoft seems unwilling to fix it.

Re-encode your DNG files in a modern software. This solves the problem.

[plastikman]

Danke Mario. I see i used the wrong queries for some of my questions.

Any idea how to create an auto-driven category based on a sub-level keyword? IMatch is very powerful and complex software so sometimes it is hard to figure out if I need to use tag based filters, formulas or even filter collections. The documentation is amazing but some things are just complex if you're not a programmer I guess. Anyway hats off especially the documentation is second to none compared to any other software I ever used.

For re-encoding the DNG files... how would that work? I converted them like ages ago probably in Lightroom 3 or something. But since they are already converted and I don't have originals any more/embedded I am not sure how I can re-encode them without quality loss etc.?

[Mario]

Any idea how to create an auto-driven category based on a sub-level keyword?

IMatch automatically groups your files by their keywords in the The @Keywords Category.
If your keywords start with the same term, they will also be grouped together in @Keywords, no?

You can also create a data-driven category based on parts keywords using the options provided.
Read the data-driven help topic to understand how data-driven categories are creates and how IMatch derives the levels from the data and settings you configure.
Note that keywords are hierarchical and files can have any number of keywords. This adds extra complexity. @Keyword deals with that already. If you really need to create your own data-driven categories on keywords, check out the "use parts of value" and "value splitting" options to try to extract individual levels from your hierarchical keywords.

Please understand that with 60-80 emails per day plus extended traffic in this community due to the IMatch 2020 release I don't have the time to sit down and figure this out for you.

I converted them like ages ago probably in Lightroom 3 or something.

I cannot be of further assistance. I have only a handful or wrongly oriented DNG files here which cause Windows WIC to take a minute (!) to rotate them. That's why I know about this effect. It was reported a couple of times years ago. Microsoft has been informed several times over the years by me and others. No effect to far.

Usually it takes WIC/DirerctX 0.1s to rotate an image. Why it takes 60s for these images, I don't know. Something weird must be in the file that causes a problem somewhere in the Microsoft DirectX libraries.
Re-encoding the images or opening them and saving them again as DNG in a software fixes this problem. This is the quickest solution.

[Carlo Didier]

For re-encoding the DNG files... how would that work? I converted them like ages ago probably in Lightroom 3 or something. But since they are already converted and I don't have originals any more/embedded I am not sure how I can re-encode them without quality loss etc.?

I don't know for Lightroom, but it should be very similar to ACR. If I open a number of DNGs in ACR, I can ask it to update the embedded preview, but you should probably first check the LR settings to select which size of preview you want (mine are always full size).

[plastikman]

Any idea how to create an auto-driven category based on a sub-level keyword?

IMatch automatically groups your files by their keywords in the The @Keywords Category.
If your keywords start with the same term, they will also be grouped together in @Keywords, no?

You can also create a data-driven category based on parts keywords using the options provided.
Read the data-driven help topic to understand how data-driven categories are creates and how IMatch derives the levels from the data and settings you configure.
Note that keywords are hierarchical and files can have any number of keywords. This adds extra complexity. @Keyword deals with that already. If you really need to create your own data-driven categories on keywords, check out the "use parts of value" and "value splitting" options to try to extract individual levels from your hierarchical keywords.

For the "Google Vision|" example I guess you're right and it could work like that although I would miss the grouping per letter since the keywords are large quantity.

For the Events | wedding, daytrip, holiday etc. example since I have "Events" checked as a group in the Thesaurus it is not written as a hierarchical keyword so I cannot drop it down in auto keywords and all the "event types" are spread out across the keyword list.

But anyway, I will experiment some more and don't feel obliged to solve it for me I hope someone else can help me out since I am pretty sure I am not the first one running into these questions.

Thanks for your input so far. Really pleased with what I am seeing.

[Mario]

I would miss the grouping per letter since the keywords are large quantity.

You can let IMatch automatically group your keywords in @Keywords by the first 1,2 ... n letters.