Method and apparatus for encoding, transmitting and decoding volumetric video
Methods, devices and streams are disclosed for encoding a depth atlas representative of the geometry of a volumetric video. Views to be encoded are analyzed to detect regions of the views with simple depth or color, that is regions for which the depth or color has a local variance lower than a given threshold. Resolution of such regions is reduced and the atlas comprises first regions in full resolution and downscaled second regions. Metadata indicating whether a patch is a downscaled region and, if so the downscaling factor, are associated with the atlas in the data stream. The decoder uses these metadata to compose the view from different patches.
1 . A method for encoding a depth view of a Multiview plus Depth frame (MVD) in an atlas of patches, a patch being a rectangular region of the depth view, the method comprising:
dividing the depth view in first rectangular regions and second rectangular regions based on complexity of geometrical information within the rectangular regions, wherein the second rectangular regions are candidates for downscaling;
downscaling a resolution of the second rectangular regions in a vertical direction by a vertical factor and in a horizontal direction by a horizontal factor to obtain downscaled second rectangular regions; and
packing the first rectangular regions and the downscaled second rectangular regions in the atlas in association with metadata indicating for each patch of the atlas whether the patch is a first rectangular region or a downscaled second rectangular region and, if so, indicating the vertical factor and the horizontal factor used for the downscaling of the resolution.
2 . The method of claim 1 , wherein the vertical and horizontal factors are different for two second rectangular regions.
3 . The method of claim 1 , wherein every pixel of each second rectangular region has a valid depth value.
4 . The method of claim 1 , further comprising:
encoding in a data stream, information indicating that patches have different resolution, the atlas, and the metadata.
5 . A non-transitory computer readable medium storing instructions which, when executed by one or more processors, cause the one or more processors to perform the method of claim 1 .
6 . The method of claim 1 , wherein the first rectangular regions are regions comprising complex geometrical information and the second rectangular regions are regions having a planar surface.
7 . A device for encoding a depth view of a Multiview plus Depth frame (MVD) in an atlas of patches, a patch being a rectangular region of the depth view, the device comprising a processor configured to perform:
dividing the depth view in first rectangular regions and second rectangular regions based on complexity of geometrical information within the rectangular regions, wherein the second rectangular regions are candidates for downscaling;
downscaling a resolution of the second rectangular regions in a vertical direction by a vertical factor and in a horizontal direction by a horizontal factor to obtain downscaled second rectangular regions; and
packing the first rectangular regions and the downscaled second rectangular regions in the atlas in association with metadata indicating for each patch of the atlas whether the patch is a first rectangular region or a downscaled second rectangular region and, if so, indicating the vertical factor and the horizontal factor used for the downscaling of the resolution.
8 . The device of claim 7 , wherein the vertical and horizontal factors are different for two second rectangular regions.
9 . The device of claim 7 , wherein every pixel of each second rectangular region has a valid depth value.
10 . The device of claim 7 , wherein the processor is further configured to encode in a data stream, information indicating that patches have different resolution, the atlas, and the metadata.
11 . The device of claim 7 , wherein the first rectangular regions are regions comprising complex geometrical information and the second rectangular regions are regions having a planar surface.
12 . A method for decoding a depth view of a Multiview plus Depth frame (MVD) from an atlas packing patches, a patch being a rectangular region of the depth view, the method comprising:
obtaining metadata indicating for each patch of the atlas whether the patch is a first patch or a second patch being a downscaled rectangular region, and, if so, indicating a vertical factor and a horizontal factor, wherein the first patch is a rectangular region comprising complex geometrical information and the second patch is a rectangular region having a planar surface;
upscaling resolution of each second patch in a vertical direction by the vertical factor and in a horizontal direction by the horizontal factor; and
composing the depth view from each first patch and each upscaled second patch.
13 . A non-transitory computer readable medium storing instructions which, when executed by one or more processors, cause the one or more processors to perform the method of claim 12 .
14 . A device for decoding a depth view of a Multiview plus Depth frame (MVD) from an atlas packing patches, a patch being a rectangular region of the depth view, the device comprising a processor configured to perform:
obtaining metadata indicating for each patch of the atlas whether the patch is a first patch or a second patch being a downscaled rectangular region, and, if so, indicating a vertical factor and a horizontal factor, wherein the first patch is a rectangular region comprising complex geometrical information and the second patch is a rectangular region having a planar surface;
upscaling resolution of each second patch in a vertical direction by the vertical factor and in a horizontal direction by the horizontal factor; and
composing the depth view from each first patch and each upscaled second patch.