Super-Resolution Enlarger for Moving Picture Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional moving picture encoding systems face inefficiencies due to insufficient code rate allocation, leading to degraded image quality and reduced encoding efficiency, particularly when encoding images with spatial frequency components exceeding standard resolution, and fail to effectively transmit additional information beyond the standard resolution.
Innovation Solution
A moving picture encoding system comprising a first encoder, a super-resolution enlarger, and a resolution converter that processes input images with standard resolution to create super-resolution enlarged images, allowing a second encoder to utilize these enhanced images for improved encoding, using decoded pictures from the first encoder as reference for prediction and encoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional encoding systems allocate code rate to standard resolution encoding, then encoding efficiency is maintained, but spatial frequency information beyond standard resolution cannot be transmitted
Solution Approach 1:
The patent segments the encoding process into two distinct layers: a base layer encoding standard resolution pictures, and an enhancement layer encoding super-resolution enlarged pictures. This segmentation allows spatial frequency information beyond standard resolution to be transmitted in the enhancement layer without compromising the efficiency of base layer encoding, thereby resolving the contradiction between information loss and encoding efficiency.
Solution Approach 2:
The patent introduces a new dimension by creating super-resolution enlarged pictures with higher spatial resolution than the original standard resolution pictures. This dimensional expansion in resolution space enables the transmission of additional spatial frequency information that would otherwise be lost, while the hierarchical structure maintains overall encoding efficiency.
2Device complexity
If code rate is insufficiently allocated, then encoding complexity is reduced, but image quality and motion vector precision deteriorate
Solution Approach 1:
The patent applies partial action by allocating code rate selectively: the base layer receives sufficient code rate to ensure acceptable image quality and encoding efficiency, while the enhancement layer receives additional code rate specifically for transmitting super-resolution information. This partial allocation approach improves image quality without unnecessarily increasing overall encoding complexity.
3Productivity
If hierarchical encoding is implemented with resolution conversion, then encoding efficiency is improved, but spatial frequency components beyond standard resolution are limited
Solution Approach 1:
The patent applies preliminary action by performing super-resolution enlargement on pictures before the encoding process. This preprocessing step creates enhanced pictures with extended spatial frequency components that can then be encoded in the enhancement layer, allowing the hierarchical encoding system to preserve and transmit spatial frequency information that would otherwise be lost during resolution conversion.
Data Source
AI summary
A first super-resolution enlarger (103) works on moving pictures input with a standard resolution, implementing a process for a super-resolution enlargement including information on frequency components in the spatial and temporal directions that has been potentially contained in the input moving pictures but unable to express to a sufficient degree by the standard resolution, and provides super-resolution enlarged signals, which are returned to the standard resolution at a first resolution converter (104). The super-resolution enlarged signals as returned to the standard resolution are encoded at a second encoder (107). A first encoder (102) encodes moving pictures input with the standard resolution, and a multiplexer (109) multiplexes a sequence of encoded bits from a first encoder (102), a sequence of encoded bits from the second encoder (107), and the like. The second encoder (107) employs local decoded signals in the first encoder (102) or processed signals thereof, as reference signals.