Depth imaging system with downsampled reflected illumination

1. A method for operating a depth imaging system, comprising:

illuminating an environment using an optical source configured to output modulated structured light comprising a pattern of dots;

receiving reflected illumination at an optical sensor comprising a 2D pixel grid;

downsampling received reflected illumination from the 2D pixel grid to a plurality of 1D dot trail vectors; and

for each dot trail vector, indicating a depth based on a time-of-flight depth measurement and a triangulation factor.

2. The method of claim 1 , wherein downsampling received reflected illumination from the 2D pixel grid to the plurality of 1D dot trail vectors is based at least on a calibration between the optical source and the optical sensor.

3. The method of claim 1 , wherein downsampling received reflected illumination from the 2D pixel grid to the plurality of 1D dot trail vectors includes downsampling depth and filtered active brightness values generated based at least on received reflected illumination for each pixel of the 2D pixel grid.

4. The method of claim 3 , further comprising:

scanning the 2D pixel grid along a scanning dimension;

recording each dot trail vector that intersects with the scanning dimension;

interpolating values between pixels that are adjacent in the scanning dimension; and

storing values for filtered active brightness and depth in a dot trail table.

5. The method of claim 1 , wherein downsampling received reflected illumination from the 2D pixel grid to the plurality of 1D dot trail vectors includes downsampling one or more of raw depth sensor measurements, complex domain depth measurements, and wrapped phase values generated based at least on received reflected illumination for each pixel of the 2D pixel grid.

6. The method of claim 1 , wherein each 1D dot trail vector comprises a plurality of samples, each sample corresponding to a specific distance between a surface of the environment and the optical sensor.

7. The method of claim 1 , further comprising:

for each sample along a 1D dot trail vector, determining a consistency error between an expected time-of-flight depth trail and the triangulation factor;

determining a sample having a minimum consistency error; and

indicating the time-of-flight depth measurement for the sample having the minimum consistency error.

8. The method of claim 1 , wherein indicating the depth is further based at least on filtered active brightness peak detection along the 1D dot trail vector.

9. The method of claim 8 , further comprising:

detecting peaks in filtered active brightness along the 1D dot trail vector;

for each detected peak, determining a consistency error between an expected time-of-flight depth trail and the triangulation factor;

determining a sample having a minimum consistency error; and

indicating a triangulated depth measurement for the sample having the minimum consistency error.

10. A depth imaging system, comprising:

an optical source configured to output modulated structured light comprising a pattern of dots;

an optical sensor comprising a 2D pixel grid;

a controller configured with instructions to:

illuminate an environment using the optical source;

receive reflected illumination at the optical sensor;

downsample received reflected illumination from the 2D pixel grid to a plurality of 1D dot trail vectors; and

for each dot trail vector, indicate a depth based on a time-of-flight depth measurement and a triangulation factor.

11. The depth imaging system of claim 10 , wherein downsampling received reflected illumination from the 2D pixel grid to the plurality of 1D dot trail vectors is based at least on a calibration between the optical source and the optical sensor.

12. The depth imaging system of claim 10 , wherein downsampling received reflected illumination from the 2D pixel grid to the plurality of 1D dot trail vectors includes downsampling depth and filtered active brightness values generated based at least on received reflected illumination for each pixel of the 2D pixel grid.

13. The depth imaging system of claim 12 , wherein the controller is further configured with instructions to:

scan the 2D pixel grid along a scanning dimension;

record each dot trail vector that intersects with the scanning dimension;

interpolate values between pixels that are adjacent in the scanning dimension; and

store values for filtered active brightness and depth in a dot trail table.

14. The depth imaging system of claim 12 , where downsampling received reflected illumination from the 2D pixel grid to a plurality of 1D dot trail vectors comprises raster scanning the 2D pixel grid.

15. The depth imaging system of claim 10 , wherein each 1D dot trail vector comprises a plurality of samples, each sample corresponding to a specific distance between a surface of the environment and the optical sensor.

16. The depth imaging system of claim 10 , wherein the controller is further configured with instructions to:

for each sample along a 1D dot trail vector, determine a consistency error between an expected time-of-flight depth trail and the triangulation factor;

determine a sample having a minimum consistency error; and

indicate the time-of-flight depth measurement for the sample having the minimum consistency error.

17. The depth imaging system of claim 10 , wherein indicating the depth is further based at least on filtered active brightness peak detection along the 1D dot trail vector.

18. The depth imaging system of claim 17 , wherein the controller is further configured with instructions to:

detect peaks in filtered active brightness along the 1D dot trail vector;

for each peak, determine a consistency error between an expected time-of-flight depth trail and the triangulation factor;

determine a sample having a minimum consistency error; and

indicate a triangulated depth measurement for the sample having the minimum consistency error.

19. A method for operating a depth imaging system, comprising:

illuminating an environment using an optical source configured to output modulated structured light comprising a pattern of dots;

receiving reflected illumination at an optical sensor comprising a 2D pixel grid;

downsampling received reflected illumination from the 2D pixel grid to a plurality of 1D dot trail vectors;

determining an estimated depth for each 1D dot trail vector;

based on the estimated depth being below a threshold depth, indicating the depth is based at least on filtered active brightness peak detection along the 1D dot trail vector; and

based on the estimated depth being above the threshold depth, indicating a depth based on a time-of-flight depth measurement and a triangulation factor.

20. The method of claim 19 , further comprising:

based on the estimated depth being below the threshold depth, detecting peaks in filtered active brightness along the 1D dot trail vector;

for each peak, determining a consistency error between the expected time-of-flight depth trail and the triangulation factor;

determining a sample having a minimum consistency error; and

indicating a triangulated depth measurement for the sample having the minimum consistency error; and

based on the estimated depth being above the threshold depth, for each sample along the 1D dot trail vector, determining the consistency error between the expected time-of-flight depth trail and the triangulation factor;

determining the sample having the minimum consistency error; and

indicating the time-of-flight depth measurement for the sample having the minimum consistency error.