Super-resolution in structured light imaging

A device comprising:

• a projector component configured to project a structured light pattern;
• an imaging sensor component configured to capture a reference image and a second image;
• and
• a memory configured to store software instructions that, when executed by at least one processor in the device, cause the at least one processor to:
• extract a first feature in the reference image and a second feature in the second image;
• determine a match between the first feature and the second feature;
• compute estimated least square parameter values for a two-dimension (2D) image alignment transformation model based on the determined match between the first feature and the second feature;
• apply the 2D image alignment transformation model with the estimated least square parameter values to the second feature to generate a transformed second feature;
• determine an error value between the transformed second feature and the first feature;
• in response to the error value being less than a threshold, rectify the second image by applying the 2D image alignment transformation model with the estimated least square parameter values to the second image;
• generate a combined image based in part on combining the rectified second image with the reference image, wherein the combined image has a higher resolution than the reference image and the second image;
• and
• generate a depth image using the combined image.