Depthmap compression using a primitive

A computer-implemented method for compressing a depthmap, the method comprising:

• grouping a plurality of pixels of a depthmap, each grouped pixel of the depthmap comprising a depth value and being associated with an optic ray aligned with a camera capturing the depthmap, each of the plurality of depth values being within a predetermined offset from one another;
• generating a primitive based on the grouped plurality of pixels, the primitive comprising an identification value, wherein each of the plurality of optic rays intersects a corresponding portion of the primitive;
• transforming the depthmap based on the primitive, wherein each grouped pixel of the transformed depthmap comprises the identification value;
• and
• compressing, by a processor, the transformed depthmap using a lossless image compression scheme.

A machine-readable medium encoded embodying executable instructions for compressing a depthmap, the instructions comprising code for:

• grouping a plurality of pixels of a depthmap, each grouped pixel of the depthmap comprising a depth value and being associated with an optic ray aligned with a camera capturing the depthmap, each of the plurality of depth values being within a predetermined offset from one another;
• generating a primitive based on the grouped plurality of pixels, the primitive comprising an identification value, wherein each of the plurality of optic rays intersects a corresponding portion of the primitive;
• transforming the depthmap based on the primitive, wherein each grouped pixel of the transformed depthmap comprises the identification value;
• compressing, by a processor, the transformed depthmap using a lossless image compression scheme;
• and
• combining metadata with the compressed transformed depthmap into a transmission file, the metadata comprising at least one of the identification value, identification information of each of the grouped plurality of pixels, and one or more parameters defining the primitive.

A computer-implemented method for decoding a compressed depthmap, the method comprising:

• receiving a compressed version of a depthmap;
• decompressing, by a processor, the compressed version of the depthmap using a lossless image decompression scheme to generate a decompressed version of the depthmap, wherein the decompressed version of the depthmap is transformed based on a primitive comprising an identification value, the primitive being generated based on a grouped plurality of pixels of an original version of the depthmap,
• wherein each grouped pixel of the original version of the depthmap comprises a depth value and is associated with an optic ray aligned with a camera capturing the original version of the depthmap, each of the plurality of depth values being within a predetermined offset from one another, each of the plurality of optic rays intersecting a corresponding portion of the primitive, and
• wherein each grouped pixel of the transformed depthmap comprises the identification value;
• and
• decoding the transformed depthmap to generate the original version of the depthmap.