Method and apparatus for bit allocation in offline video coding
View Patent ↗The method and apparatus for offline video coding, as described herein, calculates per-frame QP offset contribution from the amount of temporal masking effect at a frame, and then properly combines the calculated per-frame QP offset contribution with the original QP offset contribution from the frame type. The resulting frame-level QP offset accounts for both the frame type and temporal masking effect, and hence, is more comprehensive.
1. A method comprising the steps of:
receiving a video sequence having at least one frame;
calculating per-frame QP offset contribution from an amount of temporal masking effect at a frame (ΔQP Masking );
combining the calculated ΔQP Masking with a QP offset contribution obtained from the frame type (ΔQP Type ); and
encoding the video sequence using a bit rate obtained using the combined calculated ΔQP Masking with a QP offset contribution obtained from the frame type (ΔQP Type ).
2. The method of claim 1 , further comprising the step of:
calculating frame complexity for each frame;
selecting frame type based on GOP boundary and GOP coding pattern for each GOP;
calculating the per frame QP-offset contribution based on frame type (ΔQP Type ).
3. The method of claim 2 , further comprising the steps of:
calculating an average ΔQP Masking over all frames;
normalizing ΔQP Masking ; and
calculating a final per frame ΔQP.
4. The method of claim 3 , further comprising:
calculating an allocated bit budget and rate control using the calculated final per frame ΔQP.
5. The method of claim 1 , wherein said calculating per-frame QP offset contribution from an amount of temporal masking effect at a frame (ΔQP Masking ) is performed using ΔQP Masking (n)=a·Compl Masking (n), where a=0.5 and Compl masking (n) comprises a temporal masking complexity of a frame calculated as an average of a current frame's neighboring frames in a predetermined size of window.
6. The method of claim 2 , wherein said step of calculating frame complexity for each frame is performed using forward inter-frame prediction.
7. A computer program product comprising a non-transitory computer useable medium having a computer readable program, wherein the computer readable program when executed on a computer causes the computer to perform method steps for bit allocation in video encoding, including:
receiving a video sequence having at least one frame;
calculating per-frame QP offset contribution from an amount of temporal masking effect at a frame (ΔQP Masking );
combining the calculated ΔQP Masking with a QP offset contribution obtained from the frame type (ΔQP Type ); and
encoding the video sequence using a bit rate obtained using the combined calculated ΔQP Masking with a QP offset contribution obtained from the frame type (ΔQP Type ).
8. The computer program product as recited in claim 7 , further including
calculating frame complexity for each frame;
selecting frame type based on GOP boundary and GOP coding pattern for each GOP;
calculating the per frame QP-offset contribution based on frame type (ΔQP Type ).
9. The computer program product as recited in claim 8 , further including:
calculating an average ΔQP Masking over all frames;
normalizing ΔQP Masking ; and
calculating a final per frame ΔQP.
10. The computer program product as recited claim 9 , further including:
calculating an allocated bit budget and rate control using the calculated final per frame ΔQP.
11. The computer program product as recited claim 7 , wherein said calculating per-frame QP offset contribution from an amount of temporal masking effect at a frame (ΔQP Masking ) is performed
using ΔQP Masking (n)=a·Compl Masking (n), where a=0.5 and Compl Masking (n) comprises a temporal masking complexity of a frame calculated as an average of a current frame's neighboring frames in a predetermined size of window.
12. An encoder for encoding video sequences, the encoder comprising
means for calculating per-frame QP offset contribution from an amount of temporal masking effect at a frame (ΔQP Masking );
means for combining the calculated ΔQP Masking with a QP offset contribution obtained from the frame type (ΔQP Type ); and
means for encoding the video sequence using a bit rate obtained using the combined calculated ΔQP Masking with a QP offset contribution obtained from the frame type (ΔQP Type ).
13. The encoder of claim 12 , further comprising:
means for calculating frame complexity for each frame;
means for selecting frame type based on GOP boundary and GOP coding pattern for each GOP;
means for calculating the per frame QP-offset contribution based on frame type (ΔQP Type ).
14. The encoder of claim 13 , further comprising:
means for calculating an average ΔQP Masking over all frames;
means for normalizing ΔQP Masking ; and
means for calculating a final per frame ΔQP.
15. The encoder of claim 14 , further comprising:
means for calculating an allocated bit budget and rate control using the calculated final per frame ΔQP.
16. The encoder of claim 12 , wherein said means for calculating per-frame QP offset contribution from an amount of temporal masking effect at a frame (ΔQP Masking ) further comprises applying the following
ΔQP Masking (n)=a·Compl Masking (n), where a=0.5 and Compl Masking (n) comprises a temporal masking complexity of a frame calculated as an average of a current frame's neighboring frames in a predetermined size of window.