Method of adaptive interpolation filtering in video coding
A method and apparatus provide adaptive interpolation filter for motion compensation with fractional-pixel accuracy. Embodiments of the present invention generate interpolated reference samples at non-integer locations based on existing reference samples in a reference block by using an interpolation filter set adaptively according to pixel location and/or pixel characteristics. A current block is then encoded or decoded using a temporal predictor including the interpolated reference samples. The adaptive interpolation filter can be applied to the prediction unit (PU). In one embodiment, the interpolation filter set consists of interpolation filters having different filter lengths. An interpolation filter with a longer tap length is applied to generate interpolated reference samples farther from the block boundary and an interpolation filter with a shorter tap length is applied to generate interpolated reference samples closer to the block boundary.
1. A method of video coding in a video coding system, the method comprising:
receiving input data associated with a current block in a current picture;
determining a current motion vector (MV) associated with the current block, the current MV indicating fractional-pixel locations in a reference picture for perform motion compensation on the current block;
locating a reference block in the reference picture according to the current MV;
identifying a prediction mode of the current block;
generating interpolated reference samples at the respective fractional-pixel locations, from reference pixels of the reference block based on at least a first interpolation filter set and a second interpolation filter set, selection of a current interpolation filter set for a current one of the interpolated reference samples from multiple interpolation filter sets being based on information associated with one or a combination of a location of the current one of the interpolated reference samples, the prediction mode of the current block, a size of the current block, an accuracy of the current MV, and a characteristic of the reference pixels of the reference block, and the first interpolation filter set and the second interpolation filter set corresponding to different filter tap lengths; and
encoding or decoding the current block using a predictor including the interpolated reference samples.
2. The method of claim 1 , wherein the current block is associated with one of multiple regions partitioned from the current picture, and the information is explicitly signaled or implicitly derived for each of the multiple regions.
3. The method of claim 2 , wherein each region corresponds to one slice, one tile, one coding tree unit (CTU), one coding tree block (CTB), one coding unit (CU) or one prediction unit (PU).
4. The method of claim 1 , wherein the multiple interpolation filter sets correspond to different filter tap lengths.
5. The method of claim 1 , wherein said multiple interpolation filter sets comprises an 8-tap interpolation filter set and a 6-Tap interpolation filter set.
6. The method of claim 1 , wherein said multiple interpolation filter sets comprises an 8-tap interpolation filter set, a 6-Tap interpolation filter set and a 4-tap interpolation filter set.
7. An apparatus of video coding in a video coding system, the apparatus comprising one or more electronic circuits configured to:
receive input data associated with a current block in a current picture;
determine a current motion vector (MV) associated with the current block, the current MV indicating fractional-pixel locations in a reference picture for perform motion compensation on the current block;
locate a reference block in the reference picture according to the current MV;
identify a prediction mode of the current block;
generate interpolated reference samples at respective fractional-pixel locations from reference pixels of the reference block based on at least a first interpolation filter set and a second interpolation filter set, selection of a current interpolation filter set for a current one of the interpolated reference samples from multiple interpolation filter sets being based on information associated with one or a combination of a location of the current one of the interpolated reference samples, the prediction mode of the current block, a size of the current block, an accuracy of the current MV, and a characteristic of the reference pixels of the reference block, and the first interpolation filter set and the second interpolation filter set corresponding to different filter tap lengths;
encode or decode the current block using a predictor including the interpolated reference samples.
8. The apparatus of claim 7 , wherein the current block is associated with one of multiple regions partitioned from the current picture, and the information is explicitly signaled or implicitly derived for each of the multiple regions.
9. The apparatus of claim 8 , wherein each region corresponds to one slice, one tile, one coding tree unit (CTU), one coding tree block (CTB), one coding unit (CU) or one prediction unit (PU).
10. The apparatus of claim 7 , wherein the multiple interpolation filter sets correspond to different filter tap lengths.
11. The apparatus of claim 7 , wherein said multiple interpolation filter sets comprises an 8-tap interpolation filter set and a 6-Tap interpolation filter set.
12. The apparatus of claim 7 , wherein said multiple interpolation filter sets comprises an 8 tap interpolation filter set, a 6-Tap interpolation filter set and a 4-tap interpolation filter set.