Sunday, November 13, 2005

Multi-exposure techniques

After a bit of searching, I found two ways of doing digital multiple exposures, Option One One is to use the screen mode and underexposed images. The underexposure of each image can be done in camera, or probably better, via RAW adjustment of a correctly exposed image, for the noise reasons mentioned before. It is worth nothing that this is about one image less than suggested in the FP book, for the given stops of underexposure.

Exposures Stops 2 1 3 1.6 4 2 5 2.3 6 2.6 7 2.8 8 3

It is also possible to vary the amount of stops reduction for a given image in the sequence if you want to increase or decrease the emphasis of a particular frame. To do this, the exposure of the other frames needs to be adjusted in the opposite direction. The frames are then combined in screen mode to get to the final image. Option Two The second method doesn't underexpose the images at all. Instead, different levels of opacity are used. The background layer is left at 100%, then layer 1 is 50% opacity, layer 2 33% opacity, layer 3 25% opacity, layer 4 is 20% opacity, and so on. Each layer opacity is [1/(layer # +1)]% The blending mode is normal in this case. An alternative is to blend the images at [1/number of images]% for all the frames, then play with blending modes like screen and lighten to get something that looks about right. Of course you can vary mode and opacity to change the impact of an image as you go along - and the layer order is also up to you and can have a big impact. There are different ways of varying the frames that are shot. The FP book describes two ways - one by just shooting handheld and moving the camera slightly. The second way described is compositing images of different subjects. The sotol cactus shows a third way, of shooting at different DoFs, then combining the resulting images in an impossible to take final image. I'm sure there are other ways to vary each frame to get intriguing results.


ursula said...

Thanks! I knew the change opacity method, but the screen blending method has me a bit confused.

Screen mode "lightens the colors of underlying layers by multiplying the inverse of the selected and underlying layers. The result is a color that is the same or a lightened version of the selected layer." I know from trying it that it sort of works in creating digital multiple-exposures, yet it seems that something is not quite right about it. Sort of like the dynamics of the pictures are lost a bit.