US-20260129178-A1 - METHOD FOR DERIVING INTRA-PREDICTION MODE ON BASIS OF REFERENCE PIXEL

US20260129178A1US 20260129178 A1US20260129178 A1US 20260129178A1US-20260129178-A1

Abstract

A method and an apparatus are disclosed for video coding for deriving intra-prediction modes based on a reference pixel. In the disclosed embodiments, a video decoding device generates sub-pixel reference lines from integer-pixel reference lines. The video decoding device derives an implicit prediction mode by using the sub-pixel reference lines. Information for the implicit prediction mode includes a prediction direction and an upward prediction mode flag that indicates whether the prediction direction is upward. The video decoding device generates an intra-predictor of the current block by using the integer-pixel reference lines and the implicit prediction mode.

Inventors

Byeung Woo Jeon
Bum Yoon Kim
Jee Hwan Lee
Jin Heo
Seung Wook Park

Assignees

HYUNDAI MOTOR COMPANY
KIA CORPORATION
Research & Business Foundation Sungkyunkwan University

Dates

Publication Date: 20260507
Application Date: 20230417
Priority Date: 20220512

Claims (16)

1 . A method performed by a video decoding device for intra-predicting a current block, the method comprising: generating sub-pixel reference lines that include a top sub-pixel reference line and a left sub-pixel reference line from integer-pixel reference lines that belong to the current block and include a top integer-pixel reference line and a left integer-pixel reference line; deriving an implicit prediction mode by using the sub-pixel reference lines, wherein information on the implicit prediction mode includes a prediction direction and an upward prediction mode flag that indicates whether the prediction direction is upward; and generating an intra-predictor of the current block by using the integer-pixel reference lines and the implicit prediction mode.
2 . The method of claim 1 , further comprising: decoding an implicit prediction mode flag that indicates whether the implicit prediction mode is to be derived; and checking a value of the implicit prediction mode flag; wherein when the implicit prediction mode flag is true, the method further comprises: deriving the implicit prediction mode.
3 . The method of claim 1 , wherein deriving the implicit prediction mode comprises: applying a mapping to the top integer-pixel reference line or the top sub-pixel reference line, calculating a value of a top cost function representing a difference between a top reference line that is mapped and a left reference line, and deriving a mapping value that minimizes the value of the top cost function, as a top implicit prediction direction; and applying the mapping to the left integer-pixel reference line or the left sub-pixel reference line, calculating a value of a left cost function representing a difference between the left reference line that is mapped and the top reference line, and deriving a mapping value that minimizes the value of the left cost function, as a left implicit prediction direction; wherein the top reference line represents the top integer-pixel reference line or the top sub-pixel reference line, and the left reference line represents the left integer-pixel reference line or the left sub-pixel reference line.
4 . The method of claim 3 , wherein deriving the implicit prediction mode comprises: normalizing the value of the top cost function calculated in the top implicit prediction direction based on a height of the current block, and normalizing the value of the left cost function calculated in the left implicit prediction direction based on a width of the current block, and wherein when a minimum value between a normalized value of the top cost function and a normalized value of the left cost function is less than a predetermined threshold, the method further comprises: determining the prediction direction and the upward prediction mode flag by using the top implicit prediction direction and the left implicit prediction direction.
5 . The method of claim 4 , further comprising: when the minimum value between the normalized value of the top cost function and the normalized value of the left cost function is greater than or equal to the predetermined threshold, setting a prediction mode of the current block to planar mode.
6 . The method of claim 3 , further comprising: determining an range of the mapping based on an aspect ratio of the current block.
7 . The method of claim 1 , further comprising: determining whether an edge is present in or absent from the integer-pixel reference lines, wherein when the edge is present in both the top integer-pixel reference line and the left integer-pixel reference line, the method further comprises: deriving the implicit prediction mode.
8 . The method of claim 7 , further comprising: when the edge is absent from both the top integer-pixel reference line and the left integer-pixel reference line, aborting the deriving of the implicit prediction mode, and setting a prediction mode of the current block to DC mode.
9 . The method of claim 1 , wherein generating the intra-predictor of the current block comprises: deriving a prediction mode of the current block from the implicit prediction mode; and generating the intra-predictor by using the prediction mode.
10 . The method of claim 9 , wherein deriving the prediction mode of the current block comprises: setting the prediction mode of the current block to the prediction direction of the implicit prediction mode by referring to the upward prediction mode flag.
11 . The method of claim 9 , wherein deriving the prediction mode of the current block comprises: setting the prediction mode of the current block to a prediction mode closest to the prediction direction of the implicit prediction mode by referring to the upward prediction mode flag.
12 . The method of claim 9 , wherein deriving the prediction mode of the current block comprises: decoding an implicit prediction mode index; deriving a predetermined number of prediction modes in an order of proximity to the prediction direction of the implicit prediction mode by referring to the upward prediction mode flag; and selecting one of derived prediction modes as the prediction mode of the current block by using the implicit prediction mode index.
13 . A method performed by a video encoding device for intra-predicting a current block, the method comprising: generating sub-pixel reference lines that include a top sub-pixel reference line and a left sub-pixel reference line from integer-pixel reference lines that belong to the current block and include a top integer-pixel reference line and a left integer-pixel reference line; deriving an implicit prediction mode by using the sub-pixel reference lines, wherein information on the implicit prediction mode includes a prediction direction and an upward prediction mode flag that indicates whether the prediction direction is upward; and generating a first intra-predictor of the current block by using the integer-pixel reference lines and the implicit prediction mode.
14 . The method of claim 13 , further comprising: determining a prediction mode of the current block; generating a second intra-predictor of the current block by using the integer-pixel reference lines and the prediction mode.
15 . The method of claim 14 , further comprising: determining, by using the first intra-predictor and the second intra-predictor, an implicit prediction mode flag that indicates whether the implicit prediction mode is to be derived; and encoding the implicit prediction mode flag.
16 . A computer-readable recording medium storing a bitstream generated by a video encoding method, the video encoding method comprising: generating sub-pixel reference lines that include a top sub-pixel reference line and a left sub-pixel reference line from integer-pixel reference lines that belong to a current block and include a top integer-pixel reference line and a left integer-pixel reference line; deriving an implicit prediction mode by using the sub-pixel reference lines, wherein information on the implicit prediction mode includes a prediction direction and an upward prediction mode flag that indicates whether the prediction direction is upward; and generating an intra-predictor of the current block by using the integer-pixel reference lines and the implicit prediction mode.

Description

TECHNICAL FIELD The present disclosure relates to a method of deriving intra-prediction modes based on a reference pixel. BACKGROUND The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. Since video data has a large amount of data compared to audio or still image data, the video data requires a lot of hardware resources, including a memory, to store or transmit the video data without processing for compression. Accordingly, an encoder is generally used to compress and store or transmit video data. A decoder receives the compressed video data, decompresses the received compressed video data, and plays the decompressed video data. Video compression techniques include H.264/Advanced Video Coding (AVC), High Efficiency Video Coding (HEVC), and Versatile Video Coding (VVC), which has improved coding efficiency by about 30% or more compared to HEVC. However, since the image size, resolution, and frame rate gradually increase, the amount of data to be encoded also increases. Accordingly, a new compression technique providing higher coding efficiency and an improved image enhancement effect than existing compression techniques is required. Intra prediction to predict pixel values of the current block to be encoded utilizes pixel information within the same picture. In intra prediction, of a plurality of intra-prediction modes, an appropriate one may be selected for the features of the video and used to predict the current block. The encoder selects and uses one of the many intra-prediction modes to encode the current block. The encoder may then pass information on that mode to the decoder. HEVC technology utilizes a total of 35 intra-prediction modes for intra-prediction, including 33 angular modes that have directionality and two non-angular modes that have no directionality. However, as the spatial resolution of videos increases from 720×480 to 2048×1024 or 8192×4096, the unit size of the prediction block becomes larger and larger, which requires more intra-prediction mode varieties to be added. As illustrated in FIG. 3A, the VVC technique utilizes 67 prediction modes for intra-prediction, which are further subdivided for intra-prediction, allowing for a greater variety of prediction directions than in the prior art. In general, the image to be encoded is partitioned into Coding Units (CUs) of various shapes and sizes and then encoded in CUs. In this case, a tree structure is the information that prescribes this partitioning. The encoder transfers the tree information to the decoder, dictating how the image is divided into CUs of different shapes and sizes. In that process, the luma (Y) and chroma (Cb, Cr) images may be split into separate CUs. Alternatively, the luma and chroma images can be split into CUs of the same shape. A technique that provides the luma and chroma images with different partition structures is referred to as a chroma separate tree (CST) technique or dual tree technique. So, when the CST technique is used, the chroma image may be partitioned according to a different partitioning method than the luma image. On the other hand, a technique that provides the luma and chroma images with the same partition structure is called a single tree technique. When the single tree technique is used, the chroma image can have the same partition structure as the luma image. In the application of the CST technology, when setting the prediction mode of the luma channel and chroma channel the encoder performs a separate rate-distortion optimization (RDO) process. However, when doing the prediction mode setting of the Cb channel and Cr channel, the encoder performs the RDO process by applying the same prediction mode to both channels. When the encoder sets and encodes the intra-prediction modes according to the above, a lot of bit amount is used to encode the intra-prediction modes. Therefore, a method needs to be considered for efficiently encoding/decoding the intra-prediction modes to increase video coding efficiency and enhance video quality. DISCLOSURE Technical Problem The present disclosure seeks to provide a video coding method and an apparatus that derive prediction modes of a luma channel and a chroma channel or generate a predictor, based on reference pixels without explicitly transmitting intra-prediction mode information in the intra-prediction of the current block. Technical Solution At least one aspect of the present disclosure provides a method performed by a video decoding device for intra-predicting a current block. The method includes generating sub-pixel reference lines that include a top sub-pixel reference line and a left sub-pixel reference line from integer-pixel reference lines that belong to the current block and include a top integer-pixel reference line and a left integer-pixel reference line. The method also includes deriving an implicit prediction mode by using the sub-pixel reference lines.