Method and apparatus for video coding using de-blocking filtering based on segmentation information
A video coding method and an apparatus using a deblocking filtering based on segmentation information are disclosed. The video coding method and the apparatus perform deblocking filtering using segmentation information of an image to avoid performing filtering at the boundary of an object within the image.
1 . A method performed by a video encoding apparatus for applying deblocking filtering on a restored area of an image, the method comprising:
performing segmentation on the restored area to partition the restored area into separate objects and generating segmentation information for identifying the objects;
selecting within the restored area, a filtering boundary to which the deblocking filtering is applied;
in a P area and a Q area which form the filtering boundary, setting a P sub-area and a Q sub-area belonging to the same object using the segmentation information; and
performing the deblocking filtering for a sub-boundary between the P sub-area and the Q sub-area,
wherein the deblocking filtering is performed based on:
a length of the sub-boundary between the P sub-area and the Q sub-area, and
for each of the P sub-area and the Q sub-area, a number of samples on a line perpendicular to the sub-boundary.
2 . The method of claim 1 , wherein the segmentation information of the objects is a map in which each object is assigned a corresponding index per pixel basis.
3 . The method of claim 1 , further comprising:
obtaining the number of objects within the restored area from a high-level, wherein
generating the segmentation information partitions the restored area into as many sub-areas as the number of objects.
4 . The method of claim 1 , further comprising:
obtaining a flag indicating whether to perform the segmentation from a high-level,
wherein, when the flag is true, the segmentation is performed.
5 . The method of claim 4 , wherein, when the same segmentation method is performed respectively on the original area and the restored area, and a pixel difference as large as a predetermined threshold value occurs between the segmentation results from the original area and the restored area, the flag is set as false.
6 . The method of claim 1 , wherein the deblocking filtering is performed when the sub-boundary between the P sub-area and the Q sub-area is longer than or equal to a predetermined sample length, and when, for each of the P sub-area and the Q sub-area, a minimum value of the numbers of samples on respective lines perpendicular to the sub-boundary is larger than or equal to a predetermined threshold value.
7 . The method of claim 6 , wherein, when the sub-boundary is shorter than the predetermined sample length or when at least one of the minimum value for the P sub-area and the minimum value for the Q sub-area is smaller than the predetermined threshold value, the deblocking filtering is not performed.
8 . The method of claim 6 , wherein performing the deblocking filtering comprises:
setting a filter length of the P sub-area to one of predetermined values according to the minimum number.
9 . The method of claim 8 , wherein performing the deblocking filtering comprises:
setting a P filter length for the P sub-area to the minimum of the filter length of the P sub-area and an initialized P filter length.
10 . The method of claim 1 , wherein performing the deblocking filtering comprises:
calculating a boundary strength of the sub-boundary between the P sub-area and the Q sub-area,
wherein, when the boundary strength is larger than a predetermined threshold value, the deblocking filtering is performed on the P sub-area and the Q sub-area adjacent to the sub-boundary.
11 . The method of claim 6 , wherein performing the deblocking filtering comprises:
calculating a degree of spatial change by using samples of a first line and samples of a last line among the lines perpendicular to the sub-boundary in the P sub-area and Q sub-area; and
performing the deblocking filtering based on the degree of spatial change.
12 . The method of claim 1 , wherein the deblocking filer is one of a long filer, a short and strong filter, or a short and weak filter.
13 . A deblocking filtering method performed by a video decoding apparatus, the method comprising:
obtaining segmentation information for identifying separate objects on a restored area;
selecting, within the restored area, a filtering boundary to which deblocking filtering is applied;
in a P area and a Q area, which form the filtering boundary, setting a P sub-area and a Q sub-area belonging to the same object using the segmentation information; and
performing the deblocking filtering for a sub-boundary between the P sub-area and the Q sub-area,
wherein the deblocking filtering is performed based on:
a length of the sub-boundary between the P sub-area and the Q sub-area, and
for each of the P sub-area and the Q sub-area, a number of samples on a line perpendicular to the sub-boundary.
14 . A method for providing video data to a video decoding apparatus, the method comprising:
encoding the video data into a bitstream; and
transmitting the bitstream to the video decoding apparatus,
wherein encoding the video data comprises:
performing segmentation on the restored area to partition a restored area into separate objects and generating segmentation information for identifying the objects;
selecting, within the restored area, a filtering boundary to which deblocking filtering is applied;
in a P area and a Q area which form the filtering boundary, setting a P sub-area and a Q sub-area belonging to the same object using the segmentation information; and
performing the deblocking filtering for a sub-boundary between the P sub-area and the Q sub-area,
wherein the deblocking filtering is performed based on:
a length of the sub-boundary between the P sub-area and the Q sub-area, and
for each of the P sub-area and the Q sub-area, the number of samples on a line perpendicular to the sub-boundary.