Photo Tagging Gramplet

From Gramps
Jump to: navigation, search
The gramplet window consists of the toolbar, the selection area (center-left) and the list of tagged people (center-right).

What is Photo Tagging Gramplet

Photo Tagging Gramplet facilitates tagging people in photographs, i.e. associating regions of images with Gramps persons.

Features

  • Image preview with scrolling and zooming
  • List of persons tagged in the current photograph
  • Adding and modifying regions and associations
  • Detecting faces automatically

Installation

The Photo Tagging gramplet can be installed from within Gramps. Go to Edit -> Preferences and click the Check now button to see a list of plugins which are not installed on your computer.

Automatic detection of faces requires the following to be installed:

  • OpenCV, its dependencies and Python bindings. This library should be available in common Linux distributions. For example, Debian package python-opencv provides the Python bindings for the library. It will require the core library packages (libopencv-*), as well as python-numpy package, to be installed. If OpenCV is not found, the automatic detection feature will be unavailable, but the gramplet should otherwise function correctly.

Usage

The gramplet window consists of:

  • the toolbar
  • the selection area
  • the list of people tagged in the image

Selection area

The selection area displays the image and highlights the regions that have been associated with person in the Gramps database.

The mouse wheel can be used to zoom in and out of the image. Alternatively, you can use the corresponding toolbar buttons.

Left click inside of one the regions makes this region active. Right-click causes the context menu to be displayed.

To tag a person, select a region of the image by pressing the left mouse button in one of its corners and dragging the pointer across the desired area. A rectangular frame will mark the selected region. Use the toolbar to select an existing person or add a new person to be associated with the region.

To modify a region, first make it active by clicking inside the region. Move the mouse pointer to one of the borders or corners of the selection box. A grabbing handle will appear that you can drag to modify the shape of the box.

To change the person associated with the region, use the toolbar buttons or the context menu.

To associate a region with a different person, click within the region and use the toolbar buttons to select or add a person. In this case, the previous association is discarded.

Toolbar

The toolbar contains nine buttons that are made active in accordance with the current state of selection.

The gramplet toolbar (the exact appearance can depend on the operating system and the version of the GTK library used).

Use the toolbar to modify the associations of the current image.

Select.gif Displays the Gramps 'Select Person' dialog. Once selection is made, the person is tagged in the current box.

New.gif Creates a new person and displays the Gramps 'Person' dialog.

Clear.gif Removes the tag from the current box.

Remove.gif Removes the current box together with the tag.

Edit.gif Displays the Gramps 'Person' dialog for the person tagged in the current box.

Zoomin.gif Zooms the image in.

Zoomout.gif Zooms the image out.

Detect.gif Runs the detection algorithm.

Settings.gif Displays the gramplet settings dialog.

List of tagged people

The list contains the people tagged in the current image, as well as the selection boxes that have not yet been associated with a particular person. A thumbnail image of the corresponding region of the image is displayed in every row.

Left-click on a row in the list selects the corresponding region in the image. Further manipulations can be

Double-click on a row with a person opens the Gramps 'Person' dialog for this person. This is equivalent to pressing the 'Edit' button in the gramplet toolbar.

To do

Some of the ideas to be implemented in future versions of the gramplet:

  • Detection of faces across multiple images
  • Integrate with Picasa