How do you do that….
Take a movie.
It has 3 dimensions. X-Y and time.
Now, for every frame of that movie, take a single pixel wide column of pixels, for the sake of argument, take it from the middle of the frame.
Each of these columns is a section of spatial information and temporal information.
Now butt those scan lines up against each other along the X axis.
If you were shooting at 60 frames per second, then every second of time would be represented as 60 columns of pixels. The longer your original movie, the wider the output image is. Shoot at a faster frame rate like 1000 frames per second, and now every 1000 columns represents one second along the X axis.
which turns into this.
Note, I used a row, not a column, for this movie.
For real fun times, I take these and offset them horizontally, heal the seam, and wrap them into donuts using Photoshop>Distort>Polar Coordinates
