Video Decoding Apparatus Variable Block Size Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image compression technologies face inefficiencies when the input image size is not an integer multiple of the macroblock size, leading to difficulties in determining optimal macroblock types and prediction modes, which deteriorates encoding and decoding efficiency.
Innovation Solution
The use of variable macroblock sizes for encoding and decoding, where the size of the macroblock is adjusted based on the available block sizes to minimize data in padded areas, and borderline parts are encoded or decoded using the smallest number of bits or exempted from processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the input image is padded to become an integer multiple of macroblock size, then the image can be processed using standard macroblock-based encoding, but the encoding efficiency deteriorates due to inclusion of padded area data in bitstream
Solution Approach 1:
The image is divided into valid image regions and padded regions, with separate processing strategies applied to each segment. The encoder identifies and processes only the valid image portion, excluding padded areas from encoding operations and bitstream generation.
Solution Approach 2:
The padded region data is extracted and excluded from the encoding process. The encoder determines the valid image boundaries and processes only those regions, removing the harmful padded area data that would otherwise be included in the bitstream.
2Device complexity
If fixed macroblock size is used for all regions, then the encoding process is simple and uniform, but the compression efficiency decreases for images with non-integer macroblock sizes
Solution Approach 1:
The macroblock size is made dynamic and adaptable to the valid image region boundaries. The encoder adjusts macroblock processing to match the actual image dimensions, allowing variable-sized macroblocks or partial macroblock processing at image borders to optimize compression efficiency.
Solution Approach 2:
The macroblock size parameter is changed from a fixed value to a variable that adapts to the image dimensions. The encoder modifies macroblock processing parameters based on the valid image region size, enabling optimal compression for different image sizes.
3Ease of operation
If the macroblock type and prediction mode are determined including padded area, then the processing is straightforward, but the optimal type and mode cannot be determined leading to efficiency loss
Solution Approach 1:
The padded area is extracted and excluded from macroblock type and prediction mode determination processes. The encoder evaluates macroblock characteristics only within the valid image region, ensuring accurate type and mode selection based on actual image content.
Solution Approach 2:
Different processing quality and criteria are applied to different regions: full macroblock analysis is performed in valid image regions for accurate type determination, while padded regions are excluded from analysis entirely. This local differentiation ensures optimal type and mode selection where it matters.
Data Source
AI summary
An apparatus of decoding a video in the unit of blocks, includes: a size setting unit to decode information on a picture size of the video and information on available block sizes from a bitstream, set an initial size of a current block to be decoded to a largest block size determined by the information on the available block sizes, determine a location of the current block within a picture, identify whether the current block is fully located inside the picture, based on the picture size, the size of the current block and the location of the current block, and reset the size of the current block to a smaller size among the available block sizes, when the current block is not fully located inside the picture; and a decoding unit to decode the current block in the reset size.