Viewpoint-optimized light field display

1. A method of displaying a light field to at least one viewer of a light field display device, the light field based on a 3D model, the light field display device comprising a plurality of spatially distributed display elements, the method including the steps of:

(a) determining the viewpoints of the eyes of the at least one viewer relative to the display device;

(b) for each eye viewpoint and each of a plurality of the display elements, rendering a partial view image representing a view of the 3D model from the eye viewpoint through the display element, the partial view image comprising a set of radiance samples, each radiance sample associated with a different object point on an object plane of the display element; and

(c) displaying, via each display element, the set of partial view images rendered for that display element, by emitting an output beam based on each radiance sample, each output beam diverging from the object point associated with the radiance sample.

2. The method of claim 1 wherein step (b) comprises, for each radiance sample, determining a viewing ray from the eye viewpoint to the object point, and determining the radiance sample based on where the viewing ray intersects the 3D model.

3. The method of claim 1 wherein step (a) includes estimating a fixation depth of the at least one viewer, and setting a focus of at least one display element so that the object distance of the at least one display element corresponds to the fixation depth.

4. The method of claim 1 wherein step (a) includes estimating a gaze direction of the at least one viewer and thereby a foveal field of view associated with each eye viewpoint, and step (b) includes rendering at least part of each partial view image at a first resolution if the display element intersects the foveal field of view of the eye viewpoint, and at a second resolution otherwise, the first resolution higher than the second resolution.

5. The method of claim 1 wherein step (a) includes estimating a gaze direction of the at least one viewer and thereby a foveal field of view associated with each eye viewpoint, estimating a fixation depth of the at least one viewer, and setting a focus of at least one display element so that its object distance corresponds to the fixation depth, if the display element intersects a foveal field of view the viewer.

6. The method of claim 1 wherein step (a) includes estimating a gaze direction and fixation depth of the at least one viewer, and setting a focus of at least one display element so that its object distance corresponds to the fixation depth of the viewer fixating closest to the display element.

7. The method of claim 1 wherein step (b) includes determining, for the at least one viewer and each display element, whether the two partial view images associated with the two eye viewpoints of the viewer would overlap, and if so rendering a partial view image representing a view of the 3D model from a viewer viewpoint through the display element, and otherwise rendering, for each eye viewpoint, a partial view image representing a view of the 3D model from the eye viewpoint through the display element.

8. The method of claim 7 wherein the viewer viewpoint is selected from the group comprising the left eye viewpoint, the right eye viewpoint, a point between the eye viewpoints, and a midpoint between the eye viewpoints.

9. The method of claim 1 wherein step (a) includes estimating a gaze direction and thereby a peripheral field of the at least one viewer, and step (b) includes determining, for the at least one viewer and each display element, whether the display element is in the peripheral field of the viewer, and if so rendering a partial view image representing a view of the 3D model from a viewer viewpoint through the display element, and otherwise rendering, for each eye viewpoint, a partial view image representing a view of the 3D model from the eye viewpoint through the display element.

10. The method of claim 1 , wherein the 3D model comprises objects substantially coincident with a surface of the display device.

11. The method of claim 10 , wherein all partial view images in the set of partial view images displayed by each display element are substantially the same.

12. A method of displaying a light field to at least one viewer of a light field display device, the light field based on a 3D model, the light field display device comprising a plurality of spatially distributed display elements, the method including the steps of:

(a) determining the viewpoints of the eyes of the at least one viewer relative to the display device;

(b) for each eye viewpoint and each of a plurality of the display elements, rendering a partial view image representing a view of the 3D model from the eye viewpoint through the display element, the partial view image comprising a set of radiance samples, each radiance sample associated with a different output direction through the display element; and

(c) displaying, via each display element, the set of partial view images rendered for that display element, by emitting an output beam based on each radiance sample, each output beam substantially collimated and parallel to the output direction associated with the radiance sample.

13. The method of claim 12 wherein step (b) comprises, for each radiance sample, determining a viewing ray from the eye viewpoint and parallel to the output direction, and determining the radiance sample based on where the viewing ray intersects the 3D model.

14. The method of claim 12 wherein step (a) includes estimating a gaze direction of the at least one viewer and thereby a foveal field of view associated with each eye viewpoint, and step (b) includes rendering at least part of each partial view image at a first resolution if the display element intersects the foveal field of view of the eye viewpoint, and at a second resolution otherwise, the first resolution higher than the second resolution.

15. A method of displaying a time-varying light field to at least one viewer of a light field display device, the time-varying light field based on a 3D animation model, the light field display device comprising a plurality of spatially distributed display elements, the method including repeating, at each discrete time in a set of successive discrete times, the steps of:

(a) evaluating the 3D animation model at the discrete time;

(b) determining the viewpoints of the eyes of the at least one viewer relative to the display device;

(c) for each eye viewpoint and each of a plurality of the display elements, rendering a partial view image representing a view of the evaluated 3D model from the eye viewpoint through the display element; and

(d) displaying, via each display element, the set of partial view images rendered for that display element.