Method for frame packing in a MPEG immersive video format
A frame packing method in an MPEG immersive video format is provided. A frame packing method in association with an immersive-video encoding and decoding method efficiently arranges, on a single screen, texture information and depth information of a basic view and an additional view to increase the coding efficiency.
1. A method performed by an immersive video decoding apparatus for unpacking a pack including atlas components of an immersive video, the method comprising:
decoding a packing flag from a bitstream;
when the packing flag is true, decoding packing information from the bitstream;
decoding a subpicture or tile from the bitstream to generate the pack; and
unpacking the atlas components from the pack by using the packing information.
2. The method of claim 1 , wherein the atlas components comprise:
videos for reconstructing the immersive video, the videos including a texture video of a basic view, a depth video of a basic view, a texture video of an additional view, and a depth video of an additional view.
3. The method of claim 2 , wherein the packing information comprises:
a flag indicating a priority of a depth video, a bit depth of a texture video, a bit depth of the depth video, a null data flag, and a flag indicating a multi-view video group.
4. The method of claim 3 , wherein, when the flag indicating the priority of the depth video is not true, the pack is composed sequentially of the texture video of the basic view, the texture video of the additional view, the depth video of the basic view, and the depth video of the additional view.
5. The method of claim 3 , wherein, when the flag indicating the priority of the depth video is true, the pack is composed sequentially of the texture video of the basic view, the depth video of the basic view, the depth video of the additional view, and the texture video of the additional view.
6. The method of claim 3 , wherein when the bit depth of the depth video is 16 bits, the pack is composed by dividing the depth video into a Most Significant Bit or MSB-part video having a bit depth of 8 bits and a Least Significant Bit or LSB-part video having a bit depth of 8 bits.
7. The method of claim 6 , wherein, when the flag indicating the priority of the depth video is true, the pack is composed sequentially of the texture video of the basic view, an MSB-part upper depth video of the basic view, an LSB-part depth video of the basic view, an MSB-part depth video of the additional view, an LSB-part depth video of the additional view, and the texture video of the additional view.
8. The method of claim 3 , further comprising:
packing a 4:0:0 format of depth information into a 4:2:0 format,
wherein, when the bit depth of the depth video is less than or equal to the bit depth of the texture video, a Y channel is filled with the 4:0:0 format of depth information, and a U channel and a V channel are filled with a preset value or a downsampled value of the Y channel.
9. The method of claim 8 , wherein, when the bit depth of the depth video is greater than the bit depth of the texture video, the Y channel is filled with a Most Significant Bit or MSB-part depth video or a Least Significant Bit or LSB-part depth video equivalent to the bit depth of the texture video, and the U channel and the V channel are filled with remaining information after filling the Y channel.
10. The method of claim 3 , wherein, when the null data flag is true, the pack contains null data, and
wherein the unpacking fills a corresponding area that corresponds to the null data with a preset value.
11. The method of claim 3 , wherein, when the flag indicating the multi-view video group is true, each group of at least one or more multi-view video groups includes a basic view and an additional view for each multi-view video group, wherein the pack is composed of only the basic view and the additional view included in one multi-view video group.
12. The method of claim 2 , further comprising:
utilizing no in-loop filter when decoding subpictures corresponding to the depth video of the basic view, the texture video of the additional view, and the depth video of the additional view.
13. A method performed by an immersive video encoding apparatus for packing atlas components of an immersive video, the method comprising:
generating the atlas components from the immersive video;
obtaining a preset packing flag;
when the packing flag is true, obtaining or generating packing information; and
generating a pack by packing the atlas components based on the packing information.
14. The method of claim 13 , wherein the atlas components comprise:
videos generated from the immersive video, including a texture video of a basic view, a depth video of a basic view, a texture video of an additional view, and a depth video of an additional view.
15. The method of claim 13 , wherein the packing information comprises:
a flag indicating a priority of a depth video, a bit depth of a texture video, a bit depth of the depth video, a null data flag, and a flag indicating a multi-view video group.
16. The method of claim 13 , further comprising:
generating a bitstream by encoding the packing flag, the packing information, and the pack; and
transmitting the bitstream to an immersive video decoding apparatus.
17. The method of claim 15 , wherein the obtaining or generating of the packing information comprises:
setting the null data flag to true when the pack includes an area of null data.