Checking if 3D Point Cloud Data Follows a Standard

This patent describes a system for ensuring that compressed 3D point cloud data, used in things like virtual reality, meets specific rules about the number of points it contains, making it compatible across different devices.

What it covers

This mechanism helps ensure that 3D point cloud data follows specific rules, especially concerning the number of points. A computing device with a decoder first takes a compressed stream of video data that represents a point cloud and decodes it. Then, special circuitry reconstructs the point cloud from this decoded data. A syntax element parser then reads specific signals embedded in the compressed data, which indicate how many points are in the video's pictures. Finally, processing hardware checks if this number of points falls within the acceptable limits for the point cloud reconstructor circuitry. For example, if a VR headset receives a point cloud, this system ensures the point count won't overwhelm the headset's processing ability, preventing crashes or glitches.

What it doesn't cover

• —Does not cover conformance checks for data types other than point clouds, such as traditional 2D video or audio streams.
• —Does not cover conformance checks for aspects of point cloud data beyond the number of points, such as color accuracy or spatial resolution.
• —Does not cover point cloud coding that does not conform to the specific PCC standard mentioned in the abstract.
• —Does not cover systems where interoperability signaling is not embedded directly within the compressed bitstream itself.

The clever bit

The novelty lies in embedding specific signals (syntax elements) directly into the compressed point cloud data bitstream to explicitly state the number of points, and then automatically checking this against the hardware's known processing limits to ensure standard conformance.

Why it matters

Point clouds are essential for representing 3D environments in virtual reality, augmented reality, and autonomous vehicles. This patent helps ensure that different devices and software can understand and process point cloud data from various sources without issues. By defining how to check for standard conformance, it helps prevent compatibility problems and makes it easier for developers to create interoperable 3D applications.

Real-world examples

• 1.Virtual reality headsets displaying 3D environments
• 2.Augmented reality applications overlaying digital objects onto the real world
• 3.Autonomous vehicle perception systems processing lidar data
• 4.3D scanning and reconstruction software
• 5.Digital twin platforms for industrial design