Image Processing Device Tile Size Restriction for Line Buffer Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems require large line buffers to support the decoding of images encoded according to the VVC standard due to fixed restrictions on tile sizes, which is inefficient and resource-intensive.
Innovation Solution
Implementing an image processing device and method that restricts the horizontal size of one tile for each level by a dynamically set restriction value, allowing for smaller line buffers and efficient encoding and decoding processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the horizontal size of one tile is restricted by a fixed value in VVC encoding standard, then the decoding can be supported, but large line buffers are required which increases resource requirements
Solution Approach 1:
The patent applies dynamics by changing the tile size restriction from a fixed value to a variable that adapts to different picture sizes. The restriction value is dynamically determined based on the actual picture dimensions, allowing the system to optimize line buffer requirements while maintaining decoding support. This is achieved by setting the tile width restriction to a proportion of the picture width rather than a fixed pixel value.
Solution Approach 2:
The patent changes the parameter of tile size restriction from a fixed constant to a variable parameter that depends on picture size. By modifying the restriction value based on actual picture dimensions, the system achieves better resource utilization. The parameter change allows the tile width to scale appropriately with different resolution requirements while reducing unnecessary buffer allocations for smaller pictures.
2Adaptability or versatility
If large line buffers are used to support VVC decoding, then decoding capability is maintained, but resource efficiency decreases
Solution Approach 1:
The system dynamically adjusts the tile size restriction based on picture size, which directly impacts line buffer requirements. By making the restriction adaptive rather than fixed, the system maintains full decoding capability for all picture sizes while optimizing resource usage. Larger pictures get appropriate tile sizes for their resolution, while smaller pictures use smaller tiles that require fewer buffer resources.
Solution Approach 2:
The patent modifies the tile size parameter to be proportional to picture size rather than fixed. This parameter change enables the system to maintain decoding versatility across different resolutions while improving resource efficiency. The adaptive parameter ensures that line buffer size scales appropriately with actual content requirements rather than always allocating maximum resources.
3Manufacturing precision
If fixed tile size restriction is applied, then encoding standard compliance is achieved, but flexibility for different picture sizes is reduced
Solution Approach 1:
The patent implements dynamics by making the tile size restriction adaptive to picture dimensions. The system maintains encoding standard compliance by following the VVC framework while introducing flexibility through dynamic parameter adjustment. The tile width restriction becomes a function of picture width, allowing the system to comply with the standard's intent while adapting to specific content requirements.
Solution Approach 2:
The patent changes the restriction parameter from a fixed value to a variable that scales with picture size. This parameter modification maintains compliance with encoding standards by preserving the tile-based structure requirement while adding flexibility for different resolutions. The adaptive parameter allows proper compliance behavior across various picture sizes without sacrificing versatility.
Data Source
AI summary
The present disclosure relates to an image processing device and an image processing method that enable image processing to be performed with a smaller number of line buffers. The image processing device includes a decoding unit that decodes a bitstream to generate an image according to an encoding standard in which a horizontal size of one tile is restricted for each level by a restriction value. The restriction value is set according to a maximum value MaxLumaPs of the number of luminance picture samples, and also restricts a vertical size of one tile. The present technology can be applied to, for example, an image processing system that performs encoding and decoding according to the VVC method.