| Page (1) of 2 - 06/14/05 |
 email article |
 print
page |
|
In the previous two installments in this tutorial series, we looked at methods for using the new Spot Healing brush to fix up flaws in a subject's face. But the same technology that allows for this kind of one-click quick fix on skin can be used in much more complex images requiring the reconstruction of a background when removing an entire object from the scene. The result is similar to what you might get from the Alien Skin filter known as Smart Fill (part of the Image Doctor collection). Except, in Photoshop CS2, this sort of background reconstruction can be applied on the fly using simple brush strokes.
Before you get too far into this, mind that this technique is not perfect for all situations. Photoshop is reconstructing the background using surrounding elements and will not always do so in a way that looks "right" to the human eye. This is particularly true of backgrounds that contain definite, identifiable objects that, when reconstructed, will look "off." But for backgrounds that are composed primarily of things like trees, leaves, foliage, rocks, dirt, water and the like, the Spot Healing brush can do a fantastic job.
If you haven't done so already, you could probably benefit by going back and reading the previous two installments in this series. You can do so by clicking on the links below.
Part 1: Spot Healing Simple Blemishes
Part 2: Resurfacing Complex Images
Removing small objects
We'll begin with a simple example: several small objects that needs to be removed from the scene. In this case, I have an image of a waterfall and lake, and the objects I need to remove are the swimmers in the lake. Here's a "before and after" shot of the process.
To accomplish this, you need to work on the elements one at a time with a brush whose diameter is no larger than, say, 10 percent the width or height of your image (whichever is smaller). The technique is short, squiggly strokes setting the start point of the stroke near the background elements you want to sample. Remember, the Spot Healing brush is sampling the surrounding pixels, and part of what determines which elements will be sampled is the start point of your stroke.
I can't think of any better way to illustrate the technique than simply to show you the Spot Healing tool in action. The movie below shows the process of eliminating the people from this scene. Click the Play button to watch.
And that's going to be the technique for just about any small object you want to remove from a scene.
Wire removal
For wire removal, the problem is a bit different. A wire, after all, is going to cut across the image in a straight line and therefore leave a potentially visible streak across the image when it's removed. So let's take a look at an example that contains two such elements--a thin wire cutting across the image horizontally and a bungee cord cutting diagonally into the image. Here's a "before and after" shot of the scene.
We'll begin with the removal of the bungee cord. In order to accomplish this, we're going to need to start the healing stroke in the middle of the cord, slightly offset into the background, and work our way up in a spiral pattern. In this way, we'll not only cover the chord, but also bring in some variation from the proximity of the cord, so it doesn't look like we've simply erased something from the image.
From there, we'll begin erasing the remainder of the cord, starting at the top (where we cut the rope previously) and spiraling downward toward the subject's shoe.
Related Sites: Creative Mac , Digital Producer , Audio Video Producer , Corporate Media News , Digital Media Designer , Oceania , MacDesignPro
Related Newsletter: DMN Newsletter , Pixels Newsletter , Mac Alert Newsletter , Digital Media Net , DMNForums , Tutorial Finder
To Comment on This Article, Click HERE
Most Recent Reader Comments:
Click Here To Read All Posts
Must be Registered to Respond (Free Registration!!!, CLICK HERE)
Source:Digital Media Online.
All Rights Reserved
AV-HS450 16+ Input HD/SD Switcher w/ dual screen MultiViewer
