Morphing

Morphing is a computer-generated special effects in sound or video recordings. When morphing between transitions are calculated between two frames or two sounds.

In contrast to dissolve an image into another image through the use of targeted additional distortion is transferred during morphing. These attempt, starting from a source image to generate a realistic transition to a target image. Therefore, the typical morphing process is to select striking picture elements (eg facial features such as eyes and mouth or object edges ) in the source and target image and distort so that their contours can be made ​​to coincide. To achieve realistic as possible effects, it is important that the source and destination image does not differ too much from each other; For example, it is easier to another human face to transform a human face as a face in an iron. As the film played, these images give the impression of a continuous transformation.

Morphing is mainly used in the film industry. 1985, this technique was first announced a broad audience through the award-winning music video Cry Godley & Creme of the band ( ex 10cc ). The first film in which the technique was used, was the fantasy film Willow by Ron Howard ( 1988). Very aware of the technology was finally in 1991 by the video clip to Black or White by Michael Jackson, where the faces in an ongoing movement of whites who were turned in by African Americans and vice versa.

In the past, transitions of films by oftmaliges building and photographing a scene ( Slow - motion capture ) in conjunction with elaborate film cuts had to be generated. Due to the steady increase in computing power, one is always more passed into the film industry to develop films with the aid of digital technology ( computers and digital image processing programs). Transitions images are realized not with slow -motion, but with morphing. In the early days of the morphing simple and less realistic-looking effects such as glare were ( slow fade the RGB values ​​of the original image to the target image ) and fading (slow Hiding and showing the RGB values ​​) are used.

Further areas of applicability of images morphing outside the film industry, for example, the use of morphing in biology and chemistry, to change models, or in forensics, to search for missing persons. In the latter, old photos of these people are changed by morphing to adapt according to the appearance of these people their age.

When morphing of sounds intermediate steps are also calculated, so that the output sound slowly changed over the newly generated intermediate sounds in the new sound.

A complex morphing process consists of three steps.

Under warping refers to the twisting and distorting ( stretching or stretching ) of an image. In this case, each position of a point in the source image is assigned to a new position. This new position is dependent on the above mentioned selected, striking picture elements, which are referred to later as a reference line pairs.

The second part of Morphingalgorithmus used tweening. It is a simple linear interpolation, which transferred the position of each point in the original image to its new location. As already mentioned, the warping each point in the original image is assigned to a new position in the target image. By tweening each point in the original image is transferred to the new, calculated by the warping position. By linear interpolation, you can create an animation. It is controlled closed by a computed interval size gradually to the new location.

The third and final part of the morph consists of the cross- Dissolving or color morphing so-called. The cross- Dissolving takes place parallel to the tweening. Here, the individual RGB values ​​of each pixel are interpolated linearly, and mixed. The interpolation is performed as a function of the source and target image, i.e., during the first interpolation, the proportion of the source image is significantly higher than that of the target image. However, then the proportion of the target image increases more and more until the last interpolation only the RGB values ​​of the target image is included, and the morphing is completed.

An example: