How does SliceLapse work?

Traditional timelapse works as in the following animation. In order to create a timelapse, pictures are taken every couple seconds. These pictures are put into a video and sped up to 24-60 frames per second to create a video. Each image is displayed in its entirety for one frame of the video. This process speeds up hours into seconds.


The following animation shows how SliceLapse works. Pictures are gathered every few seconds as in traditional timelapse. Unlike timelapse however, SliceLapse splits each image into vertical slices equaling the number of images in sequence. For the first frame, the furthest left slice of the first image is taken, the second furthest left of the second image and so on. The second frame is composed of the furthest left slice of the second image, the second furthest left slice of the third image and so on. Crucially, the furthest right slice of the first image is used in the second frame. This creates a line in the frame before which is the last image in the sequence and after which is the first image in the sequence. This frame creation process is repeated until all slices have been used.


So what does this effect create?


In a normal video the x and y axises represent coordinates and the frames represent the passage of time. In a SliceLapse however, the x axis not only represents coordinates, it also represents time. Thus, for a SliceLapse, time is not only present as the video is played through the frames, it is also present along the x axis. This creates unique distortion of elements within the video. One of the best objects for SliceLapses are car headlights at night. (See the video for examples)

The following video shows examples of regular timelapse and the same footage in SliceLapse.



How SliceLapse footage is created

In order to process the images into an effective SliceLapse, several steps are applied. First the images are processed to all remain at the same level of exposure (called deflickering). This prevents one slice from being darker than others. Next all of the images are loaded into a MATLAB script I wrote for this project. This script loads all images into memory and spits out SliceLapse frames. Combining these frames into a video, a SliceLapse is born.