Light Field Technology
Light Field (LF) technology and cameras are a new way of capturing light. LF cameras store more information than normal cameras and have important use-cases in XR and holographic displays.
Light rays can point in all kind of directions. Just a few of them hit your eye. And just a few of them hit the objective of a camera. If you bring the camera or your eye in a different position different light rays that align differently then the first one’s will be captured. The whole room is filled with all kind of light rays pointing in different direction, only a couple of them are perceived by us at every moment. All these different light rays are called a light field.
Light Field Technology uses this fact to store and transmit 3d information of a scene. By capturing light rays from different directions a 3d image can be restored from this data and a 3d object can be reconstructed. So this means, we have data from an object or a scene from different directions, but we also have the depth information of this scene or object.
For Light Field Photography this means, that the focus of a photo can be changed after the scene has already been taken, in addition the perspective of the photograph can be slightly changed.
Light Field research is a very interesting research field for XR technologies.
On the one hand, it is a way of creating holograms. A couple of startups are currently using this technology to construct 3d holographic displays. See for example Looking glass and Raytrix.
The other use case in XR is the construction of depth focus. Until now it is a huge problem for VR HMDs to bring different parts of the scene in focus and out of focus. But this is how our eyes normally work. In real life we can adjust the focus of our eyes at what we want to look at, every thing else is blurry. As this is not the case for VR, the 3d scenes in VR look artificial because they are always in focus. Light field technology could solve this problem by providing the display with images that use LF depth information. The big draw back is, that for every image there are approximately 30 to 50 images that store the 3d and depth information of the image. And they all need to be rendered simultaneously. In princible that would work if our GPUs would be powerful enough. In addition, we would need working pupil tracking which is also not yet here. So we will still have a couple of years to go until we get there but the Light Field technology will definitely come into our Headsets and enhance the realness of VR experiences.