Systems and methods for video encoding acceleration in virtual, augmented, and mixed reality (xR) applications

A method, comprising:

• receiving, by an Information Handling System (IHS), a first Simultaneous Localization And Mapping (SLAM) frame and a second SLAM frame captured by a camera mounted on a Head-Mounted Device (HMD) worn by a user and wirelessly coupled to the IHS during execution of a virtual, augmented, or mixed reality (xR) application;
• extracting a first three-dimensional camera transform matrix from the first SLAM frame from the camera and extract a second three-dimensional camera transform matrix from the second SLAM frame from the camera, wherein the first camera transform matrix describes the position and orientation of the SLAM camera in the first frame and wherein the second camera transform matrix describes the position and orientation of the SLAM camera in the second frame;
• generating a two-dimensional left-eye motion matrix and a two-dimensional right-eye motion matrix from differences between the first camera transform matrix and the second camera transform matrix;
• derive a motion hint from the left-eye motion matrix and the right-eye motion matrix, wherein the motion hint comprises a candidate motion vector;
• and
• provide the motion hint to a video encoder, wherein the video encoder is configured to use the candidate motion vector to accelerate encoding of video wirelessly transmitted by the IHS to the HMD during execution of the xR application.

A hardware memory of a Head-Mounted Device (HMD) wearable by a user, the hardware memory having program instructions stored thereon that, upon execution by a processor, cause the HMD to:

• transmit, to an Information Handling System (IHS), a first Simultaneous Localization And Mapping (SLAM) frame and a second SLAM frame captured by a camera mounted on the HMD, wherein the frames are usable by the IHS to perform Simultaneous Localization and Mapping (SLAM) of the HMD during execution of a virtual, augmented, or mixed reality (xR) application, wherein the IHS:
• extracts a first three-dimensional camera transform matrix from the first SLAM frame from the camera and extract a second three-dimensional camera transform matrix from the second SLAM frame from the camera, wherein the first camera transform matrix describes the position and orientation of the SLAM camera in the first frame and wherein the second camera transform matrix describes the position and orientation of the SLAM camera in the second frame;
• generates a two-dimensional left-eye motion matrix and a two-dimensional right-eye motion matrix from differences between the first camera transform matrix and the second camera transform matrix;
• derives a motion hint from the left-eye motion matrix and the right-eye motion matrix;
• and
• receive video compressed by the IHS, at least in part, using the motion hint determined based upon the left-eye motion matrix and the right-eye motion matrix, wherein the motion hint comprises a candidate motion vector.