Out of Bounds – Eye-catching photos made easy
Including out-of-bounds effects in slide shows and presentations is bound to get the undivided attention of the audience. Gimp has simple tools to create these image effects.
Out-of-bounds (OOB) photos refer to photos post-processed so that the contents fall outside of the borders, seemingly in three-dimensional (3D) perspective. These types of photos attract a lot of attention, make a visual theme much more interesting, and often accentuate a person or an object in the photograph. With some know-how, you can create OOB photos using only open source tools. For the most part, Gimp and its filters will suffice; however, some special-purpose photos will require Inkscape.
Typically, OOB images comprise three components: the original photograph; an image border, frequently with a distorted perspective; and a coordinated background. Simulating or adding emphasis to 3D aspects is icing on the cake. The 3D qualities are achieved with various types of shadows and other effects.
An Embedded Image
Photographs suitable for this type of montage are not what you necessarily imagine them to be. For example, a halfway-exposed photo that has captured a single object is often all you need. The entire object does not even have to be captured in the photo.
The trick is to select an area in the original photograph in such a way that it contains various constituent parts of the image. Later, these constituent parts will be situated in such a way that they conceal segments of a new border that you add. Gaining experience by tinkering helps you learn how to select the best area to use, and you will find that it really does help simply to experiment on your own.
Photographs that contain very prominent parts – or at least free-standing objects – are especially well suited to the OOB technique. Figure 1 shows the photos used in this article. Frequently, the embedded photos display a strong contrast ratio, or they have been post-processed to become brighter (Figure 2), lending a distinctive boldness. The opposite also delivers results worth viewing.
One technique is always present in the creation of OOB images: cropping. Usually this term refers to cutting photos to a particular size; however, the actual meaning of the word refers to separating a part of an image from the surrounding area. Various methods can be employed to crop an object in photographs. In the following sections, I demonstrate how to crop in Gimp.
When cropping, always work on a copy of a layer so the original is preserved in the event that problems arise. To begin, you select the relevant area of the photo with the selection tool and copy it as a separate layer into the image.
As usual, Ctrl+C copies the selection to the clipboard and Ctrl+V pastes it to the image. Clicking Layer | To New Layer completes the action. You should make sure that the layer that has been generated contains all relevant parts of the image. In Gimp, this is how a transparent layer is created for use in subsequent work steps. In other programs, it may be necessary to add an alpha channel.
Now it is time to use the eraser tool to remove the areas surrounding the object. This tool lets you create a cropped area quickly. Usually you remove content, leaving the parts of the image you want intact. However, if you remove too much material from the pasted image in the transparent layer, you can recover it by holding the Alt key while "unerasing."
In many cases, it is not necessary to work out the entire design to the last detail. Precision becomes important only in places where the object will later touch the image border. At that point, even small mistakes can be noticeable.
Another area you will need to crop includes the places you will need later for perspective shadows. (See the box titled "Nifty Shadow Placement.") Normally, rough cropping suffices for this purpose. Gimp always calculates shadows at the transition between transparency and opaque areas in a layer and then blurs the details. In this way, small defects are obscured.
Nifty Shadow Placement
Gimp offers a very special function in its Filters | Light and Shadow | Perspective filter. The shadows it generates start at the object and extend according to the settings in the Perspective dialog. Shadows will look very different from drop shadows unless you use a steep angle between the light source and the object.
One drawback with this filter is that it does not have a preview because it has only been implemented as a Python script. It is difficult to guess what a parameter does by its name. Angle defines the direction of the shadow measured in the horizontal plane. Angles with fewer than 90 degrees create shadows to the right of the object. Those with greater than 90 degrees create shadows to the left.
The Relative distance of horizon parameter specifies the distance to an imaginary horizon. The documentation defines the relative distance as the distance from the baseline of the selection or layer. The height of the selection or layer counts as the unit of distance. The Relative length of shadow should not be longer than the Relative distance of horizon.
Large values for the length of the shadow can cause significant distortions. The Blur radius setting influences the edges of the shadow, and the Color setting controls the color of the shadow, whereas Opacity lets you determine how dark the areas of shadow appear. Interpolation, which is used to influence the perspective distortion of the shadow, has limited effect. Sometimes None leads to artifacts, although the Plugin will work very fast. The Linear option offers a good compromise between speed and quality.
The Allow resizing checkbox always calculates shadows completely, even if the shadows extend outside the selection and requires increasing the size of the of the shadow layer – and therefore possibly also the image size.
At this point, you should have two layers in the image, the original background and at least one partially cropped copy of the original, which should lie exactly on top of the other.
Frame styles range from very simple to very complex. Because the frame separates the content from the background of the image, it should either look very discreet or be very obvious. In any event, it should fit the subject matter.
The simplest frames are created with lines you either draw or cut out. Gimp offers numerous possibilities for creating more complex frames. (See the box titled "Realistic Frames.") The easiest solution is to simulate a frame like those used for photos printed on paper by adding a white line that takes up a small bit of the image edges. Alternatively, you can use a Polaroid variation, which features a much wider area at the lower edge (Figure 3) of the photograph. To maintain the quality of the original, you always build the frame around it, thus increasing the size of the image.
To create realistic frames, you should put a shadow on the inner side of the frame to simulate thickness. A drop shadow outside the frame marks the distance of the frame to the background. An organized interface is helpful during this process, especially when you intentionally insert errors and blurring. Alternatively, you can use the Bevel and Emboss function from the Layer Effects tools .
The perspective function lets you simulate the location of the frame in the space. If you do not have a particular frame in mind but instead want it to look like a printed photograph, then you should make sure it does not appear too flat. Adding Filters | Light and Shadow | Lighting Effects highlights the sculptural impression.
Gimp itself offers limited possibilities for creating realistic frames. Therefore, it is best to switch over to G'MIC  or use the vector graphics program Inkscape , which is also well suited for constructing 3D frames. Nevertheless, image manipulation is always the first choice when the work involves the preparation and combination of layers for the entire composition.
Another option is to take a picture of a frame and use it for the OOB design. Note that the frame should be photographed head on from the front, or the top and the middle of the frame should serve as the central orientation point. This prevents unwanted distortions that would otherwise require more work later to correct. The plugin Filters | Enhance | GimpLensfun and Filters | Distorts | Lens Distortion can be helpful with this task.
You should avoid shadows that are not to be present in the final composition. Shadows on the frame need to agree with those in the original image. Glass frames should be avoided because they almost always create reflections. Brightness and color temperature should match as closely as possible to the original image.
Frames in OOB photos only connect to parts of the image, making them easy to create. For a simple frame, first draw a frame on the transparent layer that corresponds in size and shape to the final frame.
To transform the selection to a frame, use the Select | Border or Select | Shrink function. The first method creates rounded corners for the frame, whereas the second method does not. Next, you fill in the selection and reduce it in size to the width of the frame. Afterward, the selection, which is still active, should be deleted. The results will look like Figure 4.
Buy this article as PDF
A major setback for the Linux desktop.
Improved support for GPU in virtualization.
News site for the openSUSE community falls victim to a Wordpress exploit.
The source code is available online.
One out of three virtual machines on Microsoft Azure Cloud run Linux.
The form factor of the board makes it a drop-in replacement for Raspberry Pi.
Makes it easier for customers to move workloads into container-centric applications.
SUSE’s answer to container-centric operating systems.
Linux 4.9 is the biggest release in terms of number of commits.
The latest version of the official RHEL clone is here.