If you have a few minutes, the above video gives a brief introduction to the basic features of the software, going through two common use cases; sport and entertainment.
Alternatively, below is a written outline of the key components.
Once you have logged into Caption Pro you are given the option to create a 'New Project'. You may want a project to be a specific event or job.
When you start a new project there are a number of tabs on the left hand side corresponding to various features that can be utilized. These are covered below.
This tab is for broad project settings, among other things you can name your project and set a 'watch folder'. Any images in (or that get added to) the 'watch folder' will automatically be ingested into your project to be processed. Learn more about adding and deleting images from a project in our guide: Adding and deleting Images.
This tab handles the facial recognition settings for your project.
'Set reference faces on image confirmation' is a toggle that, when enabled, will mean that confirming an image will automatically add the faces of the named people in that image, to each respective person's face recognition record in the database. If the person's record already has it's limit of 10 reference images, then nothing will happen. If this toggle is disabled, then 'confirming' an image will not add these reference faces, regardless of how many available slots they have in their record. You can still add people to your database and edit the records manually with this toggle set of 'off'.
The 'pre recognition filters are applied prior to the images being processed therefore they cannot be amended on an image to image level. These types of filters include setting a max limit on the amount of faces the software can detect in an image, or specifying which databases the software will use for facial recognition purposes.
'Post recognition Filters' are a way of limiting the names that get added to your caption. This could be done by filtering out people the software doesn't already know (guests), soloing certain people in images, or by filtering out faces from the caption based on their size or sharpness.
These can be amended on an image to image basis in the 'filters' section of the interface with an image selected.
More information about both filter types can be found in our guide: Filtering people/faces.
This tab is where you want to set up text/number recognition in your project. The feature needs to be activated prior to processing images for it to work. You also need to add a number list for the software to use as a reference. Any names and numbers in your images that are a match to those on the list will be presented as suggestions in the UI.
You can save multiple number lists in your software (for example you might want a lot of different sports teams listed) and you can use any number of active lists on a single project.
This tab lets you add a code replacement file to your project. Code replacement is a feature that allows you to apply certain words or phrases immediately into a metadata field by typing in a specific code. This enables incredibly fast application of metadata that might otherwise need to be typed out.
In order to apply a code, simply press '\' (backslash) and continue to type the relevant code. A modal will also pop up, showing possible code or result matches for you to chose between. More information about how to do this can be found in our Code Replacement guide.
This tab is where you set your project wide metadata rules. It can be treated like a template that will be stamped automatically onto all images to get processed by Caption Pro within the associated project.
This is a very important tab as it is where you set your metadata automation rules. To do this you can insert 'formulas' into individual fields by pressing the '=' button in the desired location.
An overview of the available formulas are as follows:
The 'People' formula pastes the names of the people in your image wherever the formula is added, following the rules specified when creating the formula.
The 'original value' formual brings in and maintains the current metadata that is already in the contained field. This allows you to add other information like people formulas around current metadata (say if you already have a basic template applied with another application), without overwriting it.
The 'plural' formula allows you to set multiple outputs of text dependant on the number of people in the image. For instance, you might what the output to read 'attend an event' if there are multiple people in the caption but 'attends the event' if there is only one person in the caption. You can get creative with the plural formula, another example would be to output '(L-R)' but only when there are multiple people in the image.
The 'duplicate' formula allows you to automatically insert the contents of one metadata field directly into another.
Editing / adding people to your database.
Caption Pro uses a global database of thousands of celebrities in order to help you identify who is in your image, however you may need to add new people to your own database to improve the software's accuracy.
If the software doesn't know someone or someone is being misidentified, just click on the face box of the person in question and select "Add New Person" filling in the person's information as required. When saved, the software will apply that change to any appropriate faces in the unconfirmed tab of your project.
Adding a 'Prefix' or 'Postfix' will add a title to the name of the person which can be enabled or disabled in the 'people' formulas.. more about those later though.
Once you have added a person, the first image you add of them will be used as a reference image. Any images you 'confirm' of that person will also be added to their reference images. Each person can have a maximum of 10 reference images. You can delete or add new reference faces to a person by selecting 'edit in database'.
See the 'Confirming Images' section further down the page for tips on how to make sure you have given the person the correct name.
You can delete people from your database if you have accidentally created duplicate entries for instance. This can be done by selecting the person and pressing "Delete" in the 'edit in database' modal.
All deleted people will be forgotten across the entire project - even for images in the 'confirmed' and 'exported' tabs.
More detailed information about editing people can be found in our guide to editing people.
In order to make sure you have the correct people in your images you can hit the 'search the web' button with the suggested person selected and a google image search of the suggested person's name will appear in the pop-out modal.
When you are satisfied that the people in your image are displaying correctly in the name bar you can then 'confirm' them. This is done by either right clicking on the image(s) and pressing 'confirm' or clicking the keyboard shortcut 's' with the image(s) selected. Confirming an image will move it to the 'confirmed' tab in the thumbnail viewer.
Confirming an image will add each person's faces to their corresponding reference image pool until that person has their allocation of 10 reference images filled. At which point confirming an image of someone will no longer affect their reference images.
The other purpose of confirming an image is to stop the image from being affected by newly added faces further down the project. If you have left an image in the unconfirmed section and are happy with the names, they may still be changed automatically if the software thinks the person looks more like a newly added face.
Confirmed images will still be affected by project wide metadata changes and deleted faces from your database, they just wont be affected by potentially conflicting new face information.
In order to apply the changes to the metadata of the images you must either 'export' them or 'save' the image.
If you chose to save the metadata as opposed to exporting the images, the metadata changes you make will be applied directly to the watched image in its current location. This is the quickest way to save your changes and allows for better compatibility with various different workflows.
This can be done by right clicking on the images in the unconfirmed section and selecting "confirm and save" (this will also confirm the images for you, adding the relevant reference image to any named individual's profile). It can also be done for all confirmed images by selecting 'Save all' at the bottom of the confirmed window.
The quickest way to do this on an image to image basis is by pressing the hotkey: 'shift+enter'
If your intention is to create a copy of the original images with the new information applied, or you would like to send out via FTP directly from the software the you will want to 'export' the images.
You can do this by pressing the keyboard shortcut 'e' with the necessary image(s) selected, or there is a button to export all confirmed images in the confirmed tab.
Once you have set up an export destination in a project by sending one image, you can quickly send other images to the same destination and following the same export rules by pressing the hotkey 'shift+e'
We go into more detail about these processes in our guide on Confirming and Exporting Images.
Need a bit more help?
If you have any questions, check out the relevant links throughout this page, or take a look at our FAQ page.
Alternatively if you'd like to speak to someone in person, start a conversion in one of our chat widgets or email firstname.lastname@example.org.