Point cloud compression using non-cubic projections and masks
AssigneeApple Inc.
InventorsAlexandros Tourapis, Khaled Mammou, Yeping Su, Fabrice A. Robinet, David W. Singer
Life of this patent
Estimating the term…
Abstract
A system comprises an encoder configured to compress attribute information and/or spatial for a point cloud and/or a decoder configured to decompress compressed attribute and/or spatial information for the point cloud. To compress the attribute and/or spatial information, the encoder is configured to convert a point cloud into an image based representation. Also, the decoder is configured to generate a decompressed point cloud based on an image based representation of a point cloud.
Claims21
A system, comprising:
- one or more sensors configured to capture a plurality of points that make up a point cloud, wherein respective ones of the points comprise spatial information for the point and attribute information for the point;
- and
- an encoder configured to compress the point cloud, wherein to compress the point cloud, the encoder is configured to:
- generate, for the point cloud, one or more projected images;
- pack the generated one or more projected images into one or more image frames, wherein respective sets of pixels of the one or more image frames covered by respective ones of the projected images comprise pixel values that correspond to values for the respective ones of the projected images;
- pad pixels in the one or more image frames that are not covered by the one or more projected images with a pad;
- encode the one or more image frames;
- and
- encode information indicating pixels of the one or more image frames that correspond to active regions or non-active regions of the one or more image frames, wherein the pixels that are covered by the pad are indicated as non-active regions.
A device comprising a computer system storing program instructions that, when executed, implement a decoder configured to:
- receive one or more encoded image frames comprising projected images for a compressed point cloud and padding in pixels of the or more images that are not occupied by the projected images, wherein respective sets of pixels of the one or more encoded image frames comprise pixel values that correspond to values for the respective ones of the projected images;
- decode the one or more encoded image frames, wherein less decoding resources are allocated to decoding the padded pixels of the one or more images than are allocated to decoding the projected image pixels of the one or more image frames;
- and
- generate a decompressed version of the compressed point cloud based, at least in part, on the decoded projected images.
A method, comprising:
- receiving one or more encoded image frames comprising projected images for a compressed point cloud and padding in pixels of the or more image frames that are not occupied by the projected images, wherein respective sets of pixels of the one or more encoded image frames comprise pixel values that correspond to values for the respective ones of the projected images;
- decoding the one or more encoded image frames, wherein less decoding resources are allocated to decoding the padded pixels of the one or more images than are allocated to decoding the projected image pixels of the one or more image frames;
- and
- generating a decompressed version of the compressed point cloud based, at least in part, on the decoded projected images.
Dependencies are inferred from the claim text.