Eye of the lens
 |
 |
 |
 |
 |
 |
The long method page 2
| 2. Duplicate image and save it with a different file name. | Since the anaglyph is composed of two images, it behooves you to make sure that an unshifted copy of the image is saved so that it can be used later as the right view point image in final anaglyph merge. During the development of the left point image, many layers, masks, and gray scale images will be created, and you really do not want to lose the unaltered image in the shuffle. |
| 3. Select, layer, and hide the closest object | Next, you need to find the object in the image that is the closest to your viewpoint, and cut it to a layer in the program. The idea here, is to work from the front toward the back of the viewpoint, removing objects and placing them in layers. Click here for an example. |
| 4. Complete patterns behind the removed object. | The limitation of a flat image is that information is lost due to the fact that objects in the front, cover objects in back. Excluding the possibly of analyzing shadows and other spectral data, there is no way to truly recover that lost information. Once an object is removed, you are left with an area that has no image data, and to fill that area you can simply use the clone tool to repeat what you think was in back of the removed object. Keep in mind, there is no need to completely fill the hole left by the removed foreground object with a background pattern, because you are only going to shift the foreground object a couple of pixels to the right in the final image. However, the more you fill in, the more you can shift the object. Click here for an example. |
| 5.Goto step 3 until only the ground and/or background is left. |