Git commit 71232d8156d40c1ddc5372764a189e6b3706e3f5 by Gilles Caulier. Committed on 18/09/2016 at 08:52. Pushed by cgilles into branch 'master'.
update MainWindow chapter before next 5.2.0 release M +69 -52 digikam/using-mainwindow.docbook http://commits.kde.org/digikam-doc/71232d8156d40c1ddc5372764a189e6b3706e3f5 diff --git a/digikam/using-mainwindow.docbook b/digikam/using-mainwindow.docbook index 4bb2c0f..43b4eb5 100644 --- a/digikam/using-mainwindow.docbook +++ b/digikam/using-mainwindow.docbook @@ -4,11 +4,11 @@ <sect2 id="using-mainwindow-intro"> <title>Introduction to the Main Window</title> - <para> - <mediaobject> - <imageobject><imagedata fileref="&path;using-mainwindow-preview.png" format="PNG" /></imageobject> - </mediaobject> - </para> + <para> + <mediaobject> + <imageobject><imagedata fileref="&path;using-mainwindow-preview.png" format="PNG" /></imageobject> + </mediaobject> + </para> <sect3 id="using-areasandbars"> <title>Areas and Bars</title> @@ -261,7 +261,7 @@ <para> When you delete a photograph from &digikam; with <menuchoice> - <shortcut><keycombo action="simul"><keycap>Del</keycap></keycombo> </shortcut> + <shortcut><keycombo action="simul"><keycap>Del</keycap></keycombo></shortcut> <guimenu>Item</guimenu><guimenuitem>Move to Trash</guimenuitem> </menuchoice> it will be moved from its folder on the hard disk to the internal Trash Can. @@ -278,7 +278,7 @@ <para> <mediaobject> - <imageobject><imagedata fileref="&path;using-mainwindow-albumview.png" format="PNG" /> </imageobject> + <imageobject><imagedata fileref="&path;using-mainwindow-albumview.png" format="PNG" /></imageobject> </mediaobject> </para> @@ -398,7 +398,6 @@ </sect2> - <sect2 id="using-mytagsview"> <title>Tags View</title> <para> @@ -408,9 +407,9 @@ <para> Tags can be arranged in a hierarchical tree. This allows you to organize your tags in a logical manner. You can collapse parts of the tree in the <quote>Tags</quote> list so that you can easily find the tags that you are looking for. </para> - + <para></para> - + <para> <mediaobject> <imageobject><imagedata fileref="&path;using-mainwindow-tagsview.png" format="PNG" /></imageobject> @@ -420,7 +419,7 @@ <para> When a tag is selected in the Left Sidebar, all of the images that are marked with that tag are displayed in the View Area. </para> - + <para> But before assigning and using tags you first have to define them. If you import photographs with tags assigned &digikam; will build an appropriate tag tree during import. Other than that you have to define tags by yourself. An easy access to that is the context menu shown in the screenshot above. </para> @@ -449,44 +448,44 @@ <para> To select a photograph as the tag icon, right-click on the photograph that you want to use as the tag icon and select <guilabel>Set as Tag Thumbnail</guilabel> from the context menu. Additionally you can use drag and drop to set the tag icon. Drag the images icon and drop it on the currently selected tag in the tag list. </para> - + <para> A tool for more elaborate work on big tag trees is the <link linkend="using-tagsmngr">Tags Manager</link> which you can access by clicking the <guilabel>Open Tag Manager</guilabel> button at the top of the Tags tree. </para> - + <para> In the <link linkend="keywords">Digital Asset Management (DAM)</link> chapter of this handbook you can find some useful considerations about how to build your Tags tree. </para> - + <para> It is not always easy to build a logical hierarchy from general and generic categories. You might run into a problem like this: </para> - + <blockquote><screen> Animal - Domestic Animal - Cat - Cattle - Dog - + - Wild Animal - Bird - Cat - Cheetah - Lion - Tiger - + - Zoo Animal - Bird - Cat - Cheetah - Leopard </screen></blockquote> - + <para> In this tag tree the keyword <quote>Cat</quote> appears three times. This will not cause a problem within &digikam; but there are quite a few views where the user cannot know which of the three is applied to an image because he might not be able to see the whole hierarchy of the tag. A help could be to tag the image in a case like this also with the parent tag or even the whole hierarchy but it becomes quite unwieldy, ⪚ in the image icon if you have the tags displayed there. The second <quote>Cat</quote> can be avoided by calling it <quote>Wild Cat</quote> but calling the third one <quote>Zoo Cat</quote> is a bit out of the roof, I think. And still: there are also two <quote>Cheetah</quote>! </para> - + <para> Solution: You replace the <quote>Zoo Animal</quote> branch by a single tag <quote>Zoo</quote> which you use in addition to the tag you choose from the two remaining branches. You could even put it on the top level of your hierarchy if you have photographs taken in a zoo but not showing animals. </para> @@ -498,19 +497,19 @@ <para> Before you can get the most out of &digikam;'s tagging capabilities, you must first tag your photographs. There are a few methods for that task. Once you have identified the photographs that you want to tag you can either drag and drop them onto the tag in the Tags tree or, by right-clicking on the selected photographs in the Image Window, you can use the <guilabel>Assign Tag</guilabel> menu to select the tags you wish to set. </para> - + <para> Drag and Drop: works with both sidebars provided they are showing tags of course. With the Right Sidebar it works the obvious way: you drag the tag and drop it onto the photograph or selection of photographs you want to tag. With the Left Sidebar you have to drag the photographs to the tags. You want to know why? Just try it the other way and you will see :-) </para> - + <para> Shortcuts: as mentioned earlier in this chapter you can assign keyboard shortcuts in the Tag Properties. With these you can assign or un-assign a tag to the selected photograph(s). </para> - + <para> Context menu: by right-clicking on the selected photographs and choosing <guilabel>Assign Tag</guilabel> you will see the next menu step offering the ten most recently used tags, <guilabel>Add New Tag...</guilabel>, described at the beginning of the previous <link linkend="using-managetags">article</link>, and <guilabel>More Tags...</guilabel> which leads to what is described in the next paragraph. The context menu method is obviously very useful if you use only a limited number of tags out of your whole tree for a while. </para> - + <example> <title>Tagging Image with context menu</title> <mediaobject> @@ -525,7 +524,7 @@ <para> The Right Sidebar: this is what I would call my standard method for assigning tags. Click <guilabel>Captions/Tags</guilabel> on the Right Sidebar and then the <guilabel>Tags</guilabel> tab and you will see the whole tags tree. Simply check/un-check the boxes of the tags you want to assign/un-assign and then click <guilabel>Apply</guilabel>. If you change focus without that you will be asked whether you want to apply the changes unless you disabled that confirmation either by checking <quote>Always apply changes without confirmation</quote> or by checking <quote>Do not confirm when applying changes in the right sidebar</quote> in the Miscellaneous section of <menuchoice><guimenu>Settings</guimenu> <guimenuitem>Configure digiKam...</guimenuitem></menuchoice>. More details about the other buttons and fields in the <guilabel>Tags</guilabel> tab see in <link linkend="using-sidebar-tags">&digikam; sidebar</link>. </para> - + <para> You can label a photograph with as many tags as you like. The photograph will appear when you select any of the tags that are set against it. This way you can set a tag for each person in a photograph, as well as the place the photograph was taken, the event it was taken at, &etc; </para> @@ -560,7 +559,6 @@ You may select more than one label by <keycombo action="simul">&Ctrl;<mousebutton>left</mousebutton></keycombo> click. The selected labels are connected by boolean AND, ⪚ selecting Four Star and Yellow will display only photographs that have <emphasis>both</emphasis> labels assigned. You can perform even more sophisticated searches by using the <guilabel>Filters</guilabel> section of <link linkend="using-sidebar-filters">The Right Sidebar</link>. </para> - </sect2> <sect2 id="using-mydatesview"><title>Dates View</title> @@ -605,7 +603,7 @@ <para> We talked about quite a few views already and their searching capabilities, which are all kind of specialized. The Search View now offers a more versatile and general way of searching. There are two different approaches. </para> - + <para> <mediaobject> <imageobject><imagedata fileref="&path;using-mainwindow-searchview.png" format="PNG" /> </imageobject> @@ -644,17 +642,17 @@ <para> Advanced Search tool provides an extended search form which can be used to search in specific fields of the &digikam; database in a more sophisticated way. Click the <guilabel>Advanced Search...</guilabel> button and it will come up like this: </para> - + <para> <mediaobject> - <imageobject><imagedata fileref="&path;using-mainwindow-advancedsearchtool.png" format="PNG" /> </imageobject> + <imageobject><imagedata fileref="&path;using-mainwindow-advancedsearchtool.png" format="PNG" /></imageobject> </mediaobject> </para> <para> The field labeled <quote>Find pictures that have associated all these words:</quote> is just a duplicate of the Quick Search field. </para> - + <para> Click on one of the blue categories and there will fold out an area with all the fields you can search for in that very category. Depending on the kind of data each field contains different input fields for your search, sometimes more than one type for the same field. Simplest are selection buttons, ⪚ for colors. Then you have plain fields of the <quote>The <replaceable>field content</replaceable> contains</quote> type, drop down lists and checkboxes. Many fields have two input fields allowing you to define a range, ⪚ <quote>Find pictures with a width between</quote>. And often you find a link type thing called <quote>Any</quote> which opens a drop down field with checkboxes for the possible contents of that field. </para> @@ -665,11 +663,11 @@ <imageobject><imagedata fileref="&path;using-mainwindow-advancedsearchtool2.png" format="PNG" /></imageobject> </mediaobject> </example> - + <para> - Thanks to the labeling the search fields they are pretty much self explaining. What we have to talk about is how they work together if you fill in or select more than one. On the right side of the header (blue area at the top) you find a little link <quote>Options</quote>. If you click on that one the header will change and you see four options. + Thanks to the labeling of the search fields they are pretty much self explaining. What we have to talk about is how they work together if you fill in or select more than one. On the right side of the header (blue area at the top) you find a little link <quote>Options</quote>. If you click on that one the header will change and you see four options. </para> - + <para> <itemizedlist> <listitem><para> @@ -689,15 +687,15 @@ </para></listitem> </itemizedlist> </para> - + <para> Complicated? No, good! Because the real stuff is yet to come. Have a look at the footer of the Advanced Search Tool. Here you find <guilabel>+ Add Search Group</guilabel>. If you hit this button another list with the same categories of search fields will open below the first one (scroll down in case you don't see it right away) separated by a blue header showing the same options we were just talking about. On top of them you see an underlined OR meaning that this list is connected to the first one by boolean OR. You can change that to AND by clicking on it. </para> - + <para> To explain how it works let's make another example based on one of those we had already. I want to use the second one. That means: in the first group you checked <quote>Meet Any of the following conditions</quote> and you selected Album name <quote>Holidays</quote> and color <quote>red</quote>. As we said already this will produce all pictures labeled with <quote>red</quote> out of your whole collection <emphasis>and</emphasis> the content of the Album <quote>Holidays</quote>. Now let's assume for some reason you want to exclude all rejected files and all files with a rating lower than three stars. So you open a second group, click on the underlined OR in the header in order to change it to AND, select <quote>None of these conditions are met</quote>, fold out the <quote>Picture Properties</quote>, check under <quote>Labels</quote> the red flag (rejected) and select under <quote>Rating</quote> no star (five white stars) in the first field and two stars in the second field. Note that there is a difference between <quote>no star</quote> and <quote>No Rating assigned</quote>! If you want to exclude the pictures without any rating as well you got to open another Search Group, click on the underlined OR in the header in order to change it to AND, select <quote>None of these conditions are met</quote> and select <quote>No Rating assigned</quote> in the <quote>Picture Properties</quote> category. </para> - + <para> Which brings us to the fact that you can open as many additional Search Groups as you need to formulate your query. But since the purpose of this handbook is not to open a competition in creating the most sophisticated queries I leave it to you to figure out how far you have to go or can go in this respect and rather finish this section with a few hints about the other buttons in the Advanced Search Tool. Beside the <guilabel>+ Add Search Group</guilabel> button you see <guilabel>Reset</guilabel>. This removes all additional Search Groups and clears all the fields in the remaining one. It's a good policy to use this button before starting to create a new search, in particular when you leave the Advanced Search Tool with the intention to use the Quick Search for your next query, because otherwise the search criteria remain active even if you don't see them and screw up your new search. <guilabel>Try</guilabel> carries out the search but leaves the Advanced Search Tool window open, <guilabel>OK</guilabel> does the same but closes the window. <quote>Remove Group</quote> in the header of the additional Search Groups doesn't need an explanation, I think. </para> @@ -713,7 +711,7 @@ </para> <itemizedlist> - + <listitem> <para> Duplicates: before you can have &digikam; finding duplicates the signatures (or fingerprints) have to be calculated. You can start that process with the button <guilabel>Update fingerprints</guilabel> which initiates a total rebuild of the image signatures (may take hours if you have a large collection, an estimate is 2 hours for 10'000 images). Once the fingerprints are calculated you can use <guilabel>Find duplicates</guilabel>, but it will take a long time too as it has to compare every image with any other image. So the way to go in both cases is to confine your search to certain albums and/or tags. With the Similarity threshold you can narrow down or enlarge the search result. @@ -726,7 +724,7 @@ </mediaobject> </example> </listitem> - + <listitem> <para> Image (Similar items): This is a drag&drop zone where you can drop any image to find a similar one. Drag an image from anywhere over the fuzzy search icon in the left sidebar, it will open and you drop it there or use <guilabel>Find Similar...</guilabel> from the context menu of a thumbnail in any other view. You can narrow down or enlarge the resulting selection with a threshold here as well. In the field below you can enter a name for the search and save it. In the searches list below you find your saved searches. Clicking on the title bar of that list toggles the sorting order between ascending and descending. At the bottom you find an adaptive search field which can help you to find a particular search. @@ -750,13 +748,13 @@ </mediaobject> </example> </listitem> - + </itemizedlist> </sect2> <sect2> <title>People View</title> - + <para> As long as you are not just taking pictures from machines or stars or something like that, Face Management might be an interesting feature for you (even with a machine there might be a human being operating it). In &digikam; it consists of two tasks: Face Detection and Face Recognition. </para> @@ -777,19 +775,19 @@ </listitem> </itemizedlist> </para> - + <para> The first one you would choose if you didn't scan yet or if you did with a satisfying result but added new photographs since then or if you already improved a search result, ⪚ by removing face tags which obviously don't show a face. The second you would choose if you want the images already scanned to be included in the next scan. The third is more interesting in the context of Face Recognition since <quote>unconfirmed results</quote> means face tags that don't have a name assigned to them yet. </para> - + <para> - For the task we are talking about you check <guilabel>Detect faces</guilabel>, of course. Since face detection is a time-consuming task you better <emphasis>don't</emphasis> hit <guilabel>Scan</guilabel> right away, rather <guilabel>Options</guilabel>. An additional area will fold out organized with three tabs. In <guilabel>Albums</guilabel> you can confine the scan to certain albums and/or tags. In <guilabel>Parameters</guilabel> you can try to find a balance between speed and accuracy that suits your needs. Under <guilabel>Advanced</guilabel> you will find four checkboxes. The first is explained by the text above it. The second is for face recognition, see next article. + For the task we are talking about you check <guilabel>Detect faces</guilabel>, of course. Since face detection is a time-consuming task you better <emphasis>don't</emphasis> hit <guilabel>Scan</guilabel> right away, rather <guilabel>Options</guilabel>. An additional area will fold out organized with three tabs. In <guilabel>Albums</guilabel> you can confine the scan to certain albums or tags. In <guilabel>Parameters</guilabel> you can try to find a balance between speed and accuracy that suits your needs. Under <guilabel>Advanced</guilabel> you will find two checkboxes. The first is explained by the text above it. The second is for face recognition, see next article. </para> <para> Once you have chosen your options carefully you click <guilabel>Scan</guilabel> and after a while, depending on the scope of your selection, the result will be presented in the Image Area. In the Tags list of the Left Sidebar you will see the People branch of your tag tree. You will see the whole scan result only if the topmost tag <quote>People</quote> is selected. In the tree you will see a new virtual tag called <quote>Unknown</quote> which will show all those images where faces are recognized but not yet connected to a person. If you just scanned for the first time you will find the whole result also here. </para> - + <para> <example> <title>Face Detection result</title> @@ -798,9 +796,10 @@ </mediaobject> </example> </para> - + <para> - Note that the thumbnails presented here are not showing the whole image. They show the area of the image where the scan algorithm put down a <firstterm>Face Tag</firstterm>. You can see that if you click on the thumbnail. In the preview that opens then you will see the whole image with all the face tags on them. If you don't see them, click the <inlinemediaobject> + Note that the thumbnails presented here are not showing the whole image. They show the area of the image where the scan algorithm put down a <firstterm>Face Tag</firstterm>. You can see that if you click on the thumbnail. In the preview that opens you will see the whole image with all the face tags on them. If you don't see them, click the + <inlinemediaobject> <imageobject><imagedata fileref="&path;using-mainwindow-icon-showfacetags.png" format="PNG" /></imageobject> </inlinemediaobject> button in the upper left corner of the preview. If there is more than one face tag on the image it will also be represented by more than one thumbnail. In the screenshot above this is the case with the two topmost thumbnails. @@ -811,15 +810,15 @@ <imageobject><imagedata fileref="&path;using-mainwindow-facedetectiontag.png" format="PNG" /></imageobject> </mediaobject> </para> - + <para> - I have chosen this image for the screenshot because it shows one important issue: the algorithm will find inevitably details in an image that resemble a face but are actually something else. That's what the <guilabel>Remove</guilabel> button is for. It will remove the face tag from the photograph and if it is the only face tag (left) the photograph will disappear from the scan result. + This screenshot image hav been chosen because it shows one important issue: the algorithm will find inevitably details in an image that resemble a face but are actually something else. That's what the <guilabel>Remove</guilabel> button is for. It will remove the face tag from the photograph and if it is the only face tag (left) the photograph will disappear from the scan result. </para> - + <para> If it really is a face the field labeled "Who is this?" comes into play. Here you can either type in the name of an existing People tag out of your tag tree or use the drop down function to show your tag tree and select a tag. With <guilabel>Confirm</guilabel> you can save that to the database. If it was the only face tag (left) the image will disappear from the <quote>Unknown</quote> selection and you will find it once you select either the <quote>People</quote> tag or one of the tags you assigned to one of the face tags in that image. </para> - + <para> In the context menu of the preview there are two more items related to face tags: <guilabel>Add a Face Tag</guilabel> and <guilabel>Clear all faces on this image</guilabel>. </para> @@ -829,15 +828,33 @@ <sect3> <title>Face Recognition</title> <para> - Assigning People tags to face tags is an important prerequisite to Face Recognition. Very obviously &digikam; can only recognize faces if it has something to compare with. So it's up to you to show it the ropes. + Assigning People tags to face tags is an important prerequisite to Face Recognition. Very obviously &digikam; can only recognize faces if it has something to compare with. So it's up to you to show it the ropes, which means in this case to tell &digikam;: This is Lara, this is Juan, this is Peter an so forth. For that you would, after performing a Face Detection as described in the previous article, typically select the <quote>unknown</quote> tag on the Left Sidebar, click on a thumbnail and click, if necessary, on the + <inlinemediaobject> + <imageobject><imagedata fileref="&path;using-mainwindow-icon-showfacetags.png" format="PNG" /></imageobject> + </inlinemediaobject> + (Show Face Tags) button. Now you should see the face tags of that photograph. Those which have a people (or other) tag already assigned will simply show the name of that tag. The others, showing <quote>unknown</quote> faces, will show a field and two buttons as in the screenshot of the previous article. In the field labeled "Who is this?" you can either type in the name of an existing People tag out of your tag tree or use the drop down function to show your tag tree and select a tag. With <guilabel>Confirm</guilabel> you can save that to the database. </para> - - <para> TODO </para> <para> If a face tag is confirmed and thus showing only the name of the tag but not the buttons to remove, edit or confirm it, but you need this buttons because ⪚ it's wrong and you want to edit or remove it, just double-click on the name and the buttons will re-appear. </para> + <para> + Once you have a tag assigned to a few photographs you can have &digikam; looking if it can find more photographs showing the same face. To prepare that process you click <guilabel>Scan collection for faces</guilabel> but his time you select <guilabel>Recognize faces</guilabel>. + </para> + + <para> + Face Recognition is faster than Face Detection but it still makes sense to click <guilabel>Options</guilabel> and confine the scan to certain albums or tags in the <guilabel>Albums</guilabel> tab, ⪚ to the "Unknown" tag. In the <guilabel>Parameters</guilabel> tab we can play with the balance between speed and accuracy. Under <guilabel>Advanced</guilabel> you will find two checkboxes. The first is explained by the text above it. + </para> + + <para> + In case of unsatisfying results it might be helpful to use <guilabel>Clear and rebuild all training data</guilabel>. One reason can be that there are to many face tags assigned to a person which shows this person in a way that doesn't really help the search algorithm, ⪚ with sunglasses, blurred, unusual colors, carnival make up, dark shaded areas in the face, baby/kid/adult photographs mixed... Another reason to use that option can be false face recognition due to a wrong accuracy setting in the Parameters tab. + </para> + + <para> + To start the Face Recognition you click <guilabel>Scan</guilabel>. The process will tag every recognized face with the appropriate People tag out of your tag tree and the corresponding thumbnail will disappear from the "Unknown" tag selection. + </para> + </sect3> </sect2>
