Apparatuses and methods for encoding and decoding a panoramic video signal
AssigneeHuawei Technologies Co., Ltd.
InventorSergey Yurievich Ikonin
Life of this patent
Estimating the term…
Abstract
The disclosure relates to an apparatus for encoding a video signal, wherein the video signal is a two-dimensional projection of a panoramic video signal and comprises a plurality of successive frames, including a reference frame and a current frame, wherein each frame of the plurality of successive frames comprises a plurality of video coding blocks and wherein each video coding block comprises a plurality of pixels.
Claims16
An encoding apparatus for encoding a video signal, the video signal being a two-dimensional (2D) projection of a panoramic video signal and comprising a plurality of successive frames including a reference frame and a current frame, each frame of the plurality of successive frames comprising a plurality of video coding blocks, each video coding block comprising a plurality of pixels, the encoding apparatus comprising computer hardware implementing a plurality of units, including:
- an inter prediction unit for generating a predicted video coding block for a current video coding block of the current frame based on a corresponding video coding block of the reference frame, wherein the inter prediction unit is configured to:
- determine a three-dimensional (3D) reference motion vector based on a first 3D position defined by a projection of a first pixel of the current video coding block of the current frame onto a viewing sphere and based on a second 3D position defined by a projection of a second pixel of the corresponding video coding block of the reference frame onto the viewing sphere, and
- predict at least one pixel of the current video coding block based on the 3D reference motion vector;
- and
- an encoding unit configured to encode the current video coding block of the current frame based on the predicted video coding block,
- wherein the inter prediction unit is configured to predict the at least one pixel of the current video coding block based on the 3D reference motion vector by:
- selecting a further pixel from the corresponding video coding block of the reference frame;
- generating a modified 3D reference motion vector based on the 3D reference motion vector, wherein the modified 3D reference motion vector and the 3D reference motion vector have a same length, but different orientations;
- adding the modified 3D reference motion vector to a third 3D position defined by a projection of the further pixel from the corresponding video coding block of the reference frame onto the viewing sphere to generate a fourth 3D position;
- and
- determining a 2D position of the at least one pixel of the current video coding block by de-projecting the fourth 3D position onto the current frame.
A method for encoding a video signal, the video signal being a two-dimensional (2D) projection of a panoramic video signal and comprising a plurality of successive frames including a reference frame and a current frame, each frame of the plurality of successive frames comprising a plurality of video coding blocks, each video coding block comprising a plurality of pixels, the encoding method comprising:
- determining a three-dimensional (3D) reference motion vector based on a first 3D position defined by a projection of a first pixel of a current video coding block of the current frame onto a viewing sphere and based on a second 3D position defined by a projection of a second pixel of a corresponding video coding block of the reference frame onto the viewing sphere;
- predicting at least one pixel of the current video coding block based on the 3D reference motion vector to generate a predicted video coding block for the current video coding block of the current frame based on the corresponding video coding block of the reference frame;
- and
- encoding the current video coding block of the current frame based on the predicted video coding block,
- wherein predicting the at least one pixel of the current video coding block based on the 3D reference motion vector comprises:
- selecting a further pixel from the corresponding video coding block of the reference frame;
- generating a modified 3D reference motion vector based on the 3D reference motion vector, wherein the modified 3D reference motion vector and the 3D reference motion vector have a same length, but different orientations;
- adding the modified 3D reference motion vector to a third 3D position defined by a projection of the further pixel from the corresponding video coding block of the reference frame onto the viewing sphere to generate a fourth 3D position;
- and
- determining a 2D position of the at least one pixel of the current video coding block by de-projecting the fourth 3D position onto the current frame.
A decoding apparatus for decoding a video signal, the video signal being a two-dimensional (2D) projection of a panoramic video signal and comprising a plurality of successive frames including a reference frame and a current frame, each frame of the plurality of successive frames comprising a plurality of video coding blocks, each video coding block comprising a plurality of pixels, the decoding apparatus comprising computer hardware implementing a plurality of units, including:
- a decoding unit configured to decode the video signal to provide a residual video coding block associated with a current video coding block of the current frame;
- an inter prediction unit for generating a predicted video coding block for a current video coding block of the current frame based on a corresponding video coding block of the reference frame, wherein the inter prediction unit is configured to:
- obtain a three-dimensional (3D) reference motion vector based on a first 3D position defined by a projection of a first pixel of the current video coding block of the current frame onto a viewing sphere and a second 3D position defined by a projection of a second pixel of the corresponding video coding block of the reference frame onto the viewing sphere, and
- predict at least one pixel of the current video coding block based on the 3D reference motion vector;
- and
- a restoration unit configured to restore the current video coding block based on the residual video coding block and the predicted video coding block,
- wherein the inter prediction unit is configured to predict the at least one pixel of the current video coding block based on the 3D reference motion vector by:
- selecting a further pixel from the corresponding video coding block of the reference frame;
- generating a modified three-dimensional reference motion vector based on the 3D reference motion vector, wherein the modified 3D reference motion vector and the 3D reference motion vector have the same length, but different orientations;
- adding the modified 3D reference motion vector to a third 3D position defined by a projection of the further pixel from the corresponding video coding block of the reference frame onto the viewing sphere to generate a fourth three-dimensional position;
- and
- determining a 2D position of the at least one pixel of the current video coding block by de-projecting the fourth 3D position onto the current frame.
A method for decoding a video signal, the video signal being a two-dimensional (2D) projection of a panoramic video signal and comprising a plurality of successive frames including a reference frame and a current frame, each frame of the plurality of successive frames comprising a plurality of video coding blocks, each video coding block comprising a plurality of pixels, the decoding method comprising:
- decoding the video signal for providing a residual video coding block associated with a current video coding block of the current frame;
- obtaining a three-dimensional (3D) reference motion vector based on a first 3D position defined by a projection of a first pixel of the current video coding block of the current frame onto a viewing sphere and a second 3D position defined by a projection of a second pixel of a corresponding video coding block of the reference frame onto the viewing sphere;
- predicting at least one pixel of the current video coding block based on the 3D reference motion vector to generate a predicted video coding block for the current video coding block of the current frame based on the corresponding video coding block of the reference frame;
- and
- restoring the current video coding block based on the residual video coding block and the predicted video coding block,
- wherein predicting the at least one pixel of the current video coding block based on the 3D reference motion vector comprises:
- selecting a further pixel from the corresponding video coding block of the reference frame;
- generating a modified three-dimensional reference motion vector based on the 3D reference motion vector, wherein the modified 3D reference motion vector and the 3D reference motion vector have the same length, but different orientations;
- adding the modified 3D reference motion vector to a third 3D position defined by a projection of the further pixel from the corresponding video coding block of the reference frame onto the viewing sphere to generate a fourth three-dimensional position;
- and
- determining a 2D position of the at least one pixel of the current video coding block by de-projecting the fourth 3D position onto the current frame.
Dependencies are inferred from the claim text.