High throughput residual coding for a transform skipped block for CABAC in HEVC
A system utilizing a high throughput residual coding mode for CABAC in HEVC is described. The system may include an electronic device configured to obtain a bitstream; recover a binary symbol from the obtained bitstream; determine whether the binary symbol is to be decoded using a high throughput residual coding mode; in response to determining that the binary symbol is not to be decoded using the high throughput residual coding mode, use a first coding technique to obtain a block of Transformed and Quantized Coefficients (TQCs); and, in response to determining that the binary symbol is to be decoded using the high throughput residual coding mode, use a second different coding technique to obtain a residual sample.
1. A system, comprising:
a memory and a processor coupled to the memory, the processor configured to:
obtain a bitstream corresponding to a decoding block;
parse a first symbol from the bitstream, the first symbol indicates whether or not residual data of the decoding block is transformed;
parse a second symbol from the bitstream using a first coding technique when the first symbol indicates the residual data is transformed, wherein the second symbol is used to decode Transformed and Quantized Coefficients of the decoding block; and
parse a third symbol from the bitstream using a second coding technique when the first symbol indicates the residual data is not transformed, wherein the third symbol is used to decode residual samples of the decoding block, wherein the second coding technique has a higher throughput than the first coding technique, and wherein the first coding technique and the second coding technique are based on Context Adaptive Binary Arithmetic Coding (CABAC),
wherein the first coding technique comprises a Golomb-Rice (G-R) coding method to code an Absolute−3 value, and wherein the G-R coding method comprises initializing a Rice parameter at a predefined value based on at least color information (luma/chroma).
2. The system of claim 1 , wherein only the first coding technique of the first and second coding techniques comprises an Absolute−3 coding and the second coding technique does not comprise the Absolute−3 coding.
3. The system of claim 2 , wherein the second coding technique comprises an Absolute−1 or Absolute−2 coding.
4. The system of claim 1 , wherein the first coding technique comprises coding a Greater_than_1 flag and a Greater_than_2 flag, and wherein the second coding technique codes neither the Greater_than_1 flag nor the Greater_than_2 flag.
5. The system of claim 1 , wherein the second coding technique comprises the G-R coding method to code an Absolute−1 or Absolute−2 value.
6. A method of decoding an image, comprising:
obtaining a bitstream corresponding to a decoding block;
parsing a first symbol from the bitstream, the first symbol indicates whether or not residual data of the decoding block is transformed;
parsing a second symbol from the bitstream using a first coding technique when the first symbol indicates the residual data is transformed, wherein the second symbol is used to decode Transformed and Quantized Coefficients of the decoding block; and
parsing a third symbol from the bitstream using a second coding technique when the first symbol indicates the residual data is not transformed, wherein the third symbol is used to decode residual samples of the decoding block, wherein the second coding technique has a higher throughput than the first coding technique, and wherein the first coding technique and the second coding technique are based on Context Adaptive Binary Arithmetic Coding,
wherein the first coding technique comprises a Golomb-Rice (G-R) coding method to code an Absolute−3 value, and wherein the G-R coding method comprises initializing a Rice parameter at a predefined value based on at least color information (luma/chroma).
7. The method of claim 6 , wherein only the first coding technique of the first and second coding techniques comprises an Absolute−3 coding and the second coding technique does not comprise the Absolute−3 coding.
8. The method of claim 7 , wherein the second coding technique comprises an Absolute−1 or Absolute−2 coding.
9. The method of claim 6 , wherein the first coding technique comprises coding a Greater_than_1 flag and a Greater_than_2 flag, and wherein the second coding technique codes neither the Greater_than_1 flag nor the Greater_than_2 flag.
10. The method of claim 6 , wherein the second coding technique comprises the G-R coding method to code an Absolute−1 or Absolute−2 value.
11. A system, comprising:
a memory and a processor coupled to the memory, the processor configured to:
obtain a bitstream corresponding to a decoding block;
parse a first symbol from the bitstream, the first symbol indicates whether or not residual data of the decoding block is transformed;
parse a second symbol from the bitstream using a first coding technique when the first symbol indicates the residual data is transformed, wherein the second symbol is used to decode Transformed and Quantized Coefficients of the decoding block, wherein the first coding technique is based on Context Adaptive Binary Arithmetic Coding, wherein the first coding technique comprises decoding a first amount of level code flags of a syntax element associated with the decoding block and decoding a second amount of level code flags of the syntax element associated with the decoding block when the decoding block has a predefined number of Transformed and Quantized Coefficients which are greater than one, wherein the first amount of level code flags at least comprises a Greater_than_1 flag and a Greater_than_2 flag, and wherein the second amount of level code flags does not comprise the Greater_than_1 flag or the Greater_than_2 flag; and
parse a third symbol from the bitstream using a second coding technique when the first symbol indicates the residual data is not transformed, wherein the third symbol is used to decode residual samples of the decoding block, wherein the second coding technique has a higher throughput than the first coding technique,
wherein the first coding technique comprises a Golomb-Rice (G-R) coding method to code an Absolute−3 value, and wherein the G-R coding method comprises initializing a Rice parameter at a predefined value based on at least color information (luma/chroma).
12. The system of claim 11 , wherein the second coding technique comprises the G-R coding method to code an Absolute−1 value or an Absolute−2 value.