As Light Field Photography with a Hand-held Plenoptic Camera by Ng et al. demonstrated, capturing multiple images over a plane orthogonal to the optical axis enables achieving complex effects using very simple operations like shifting and averaging. In this project, we simulate depth refocusing and aperture adjustment using light field data downloaded from Stanford Light Field Archive to get some hands on experiement with lightfield camera.
Depth refocusing can be achieved by averaging shifted pictures. As we know, the objects which are far away from the camera do not vary their position significantly when the camera moves around while keeping the optical axis direction unchanged. The nearby objects, on the other hand, vary their position significantly across images. We first find the central camera coordinates using all the pictures from the grid, and then calculate the coordinates offset for all the pictures. When it comes to generate focused pictures at different depth, we shift each picture by the offset multiplied by a constant, with negative constant simulating forward focus and positive constant simulating backward focus, and then average all the pictures.
| Chess | Lego Knights |
|---|---|
|
|
Aperture adjustment can be achieved by averaging different radius circle patches of pictures from the center camera. Assume the central camera has the smallest aperture and since the whole picuture is in focus, we can assume that all the light rays go into the lense are perpendicular to the optical axis. As we increase the aperture, there will be more light rays goes into the lense, and some light rays that are far away from the center on the edge may not be perpendicular and thus creating a blurring effect. Starting from the center camera at [8, 8], we simulate the aperture adjustment by finding all the picutures that are in range of [0, 2, 4, 6, 8] from the center camera and average them.
| Chess | Lego Knights |
|---|---|
 |
 |