Depth mapping and parallel distortion correction for mixed reality

1. A mixed reality system, comprising:

a computer system configured to generate a virtual reality video stream;

a head mounted device communicatively coupled to the computer system; the head mounted device including:

a display;

a depth sensor;

and

a stereoscopic camera system;

and

a video processing device communicatively coupled to the computer system and the head mounted device, the video processing device configured to:

generate a first depth map based on time-of-flight measurements detected by the depth sensor;

generate a second depth map based on disparity mapping from stereo imagery detected by the stereoscopic camera system;

determine confidence values for respective pixel locations in the first depth map and the second depth map based on texture recognition;

blend the first depth map and the second depth map into a combined depth map based on the confidence values of the respective pixel locations in the first depth map and the second depth map;

and

combine the virtual reality video stream and the stereo imagery detected by the stereoscopic camera system based on the combined depth map and depth criteria encoded into the virtual reality video stream.

2. The mixed reality system of claim 1 , wherein the video processing device is coupled to or integrated within the head mounted device.

3. The mixed reality system of claim 1 , wherein the video processing device is further configured to:

downsample the stereo imagery before generating the second depth map; and

upsample the combined depth map to a resolution of the virtual reality video stream before combining the virtual reality video stream with the stereo imagery detected by the stereoscopic camera system.

4. The mixed reality system of claim 1 , wherein the video processing device is further configured to merge the combined depth map with optical cues from the stereo imagery detected by the stereoscopic camera system for accurate extraction of identified foreground objects based on distance.

5. The mixed reality system of claim 1 , wherein the computer system is configured to encode the depth criteria into the virtual reality video stream, and the video processing device is configured to decode the depth criteria from the virtual reality video stream, thereby enabling use of commercial-off-the-shelf (COTS) virtual reality hardware and software without modification.

6. The mixed reality system of claim 1 , wherein the video processing device is further configured to perform distortion correction on the stereo imagery before combining the virtual reality video stream with the stereo imagery detected by the stereoscopic camera system.

7. The mixed reality system of claim 6 , wherein the video processing device is configured to perform the distortion correction with a distortion corrector including:

a sparse matrix table including pixel-by-pixel coordinate mappings decimated by a selected rate;

a sparse matrix table interface configured to fetch a subset of the coordinate mappings from the sparse matrix table for interpolation;

an address bi-linear interpolator configured to calculate one or more input image coordinates for one or more pixels; and

a pixel buffer configured to buffer a subset of pixels for an output image, fetch the one or more pixels based on the one or more input image coordinates, and interpolate the one or more pixels to produce one or more output image pixels.

8. The mixed reality system of claim 7 , wherein the one or more input image coordinates for the one or more pixels are fractional coordinates.

9. The mixed reality system of claim 7 , wherein each output image pixel is based on at least four input image pixels.

10. A video processing device for a mixed reality system, the video processing device comprising a controller configured to:

generate a first depth map based on time-of-flight measurements detected by a depth sensor of a head mounted device;

generate a second depth map based on disparity mapping from stereo imagery detected by a stereoscopic camera system of the head mounted device;

determine confidence values for respective pixel locations in the first depth map and the second depth map based on texture recognition;

blend the first depth map and the second depth map into a combined depth map based on the confidence values of the respective pixel locations in the first depth map and the second depth map;

and

combine a virtual reality video stream with the stereo imagery detected by the stereoscopic camera system based on the combined depth map and depth criteria encoded into the virtual reality video stream.

11. The video processing device of claim 10 , wherein the controller is further configured to:

downsample the stereo imagery before generating the second depth map;

and

upsample the combined depth map to a resolution of the virtual reality video stream before combining the virtual reality video stream with the stereo imagery detected by the stereoscopic camera system.

12. The video processing device of claim 10 , wherein the controller is further configured to merge the combined depth map with optical cues from the stereo imagery detected by the stereoscopic camera system for accurate extraction of identified foreground objects based on distance.

13. The video processing device of claim 10 , wherein the depth criteria is encoded into the virtual reality video stream by a communicatively coupled computer system, and the controller is configured to decode the depth criteria from the virtual reality video stream after receiving the virtual reality video stream from the computer system, thereby enabling use of commercial-off-the-shelf (COTS) virtual reality hardware and software without modification.

14. The video processing device of claim 10 , wherein the controller is further configured to perform distortion correction on the stereo imagery before combining the virtual reality video stream with the stereo imagery detected by the stereoscopic camera system.

15. The video processing device of claim 14 , wherein the controller is configured to perform the distortion correction with a distortion corrector including:

a sparse matrix table including pixel-by-pixel coordinate mappings decimated by a selected rate;

a sparse matrix table interface configured to fetch a subset of the coordinate mappings from the sparse matrix table for interpolation;

an address bi-linear interpolator configured to calculate one or more input image coordinates for one or more pixels;

and

a pixel buffer configured to buffer a subset of pixels for an output image, fetch the one or more pixels based on the one or more input image coordinates, and interpolate the one or more pixels to produce one or more output image pixels.

16. The video processing device of claim 15 , wherein the one or more input image coordinates for the one or more pixels are fractional coordinates, and wherein each output image pixel is based on at least four input image pixels.

17. The video processing device of claim 15 , wherein the pixel buffer is configured with a dual-bank, virtual addressing scheme that enables multiple independent neighborhood fetches of pixels in parallel.