Dynamic Reference Frame Allocation for H.264 Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
H.264 universal decoders require a large dual data rate (DDR) memory footprint due to the need for storing multiple reference frames and co-located macroblock information, especially in High Definition (HD) decoding, which is not feasible in many systems and becomes more severe in multichannel systems.
Innovation Solution
The method configures memory for each group of pictures (GOP) to use less than the maximum number of reference frames needed, optimizing memory usage based on the specific GOP structure and stream properties, allowing the decoder to operate with only the required number of reference frames and macroblock information buffers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the decoder stores the maximum number N of reference frames in DDR memory, then the decoder can handle any GOP structure and resolution, but the memory footprint becomes excessively large (24.9 Mbytes for HD)
Solution Approach 1:
The patent implements dynamic memory allocation where the number of reference frames stored in DDR memory is adjusted based on the actual GOP structure being decoded. Instead of statically allocating memory for the maximum N frames, the system dynamically determines the actual number M of reference frames needed for the current GOP and configures memory accordingly, reducing memory footprint while maintaining decoding capability.
Solution Approach 2:
The patent changes the parameter of reference frame count from a fixed maximum value N to a variable value M that adapts to different GOP structures. By analyzing the GOP structure beforehand and determining the actual number of reference frames required, the system optimizes memory allocation parameters to match actual needs rather than worst-case scenarios.
2Adaptability or versatility
If the decoder allocates memory for N reference frames and co-located macroblock information, then all H.264 streams can be decoded, but many systems do not have this much free memory available
Solution Approach 1:
The patent applies partial action by allocating only the necessary portion of memory required for the current GOP structure rather than the full maximum allocation. By determining the actual number M of reference frames needed (which is often less than the maximum N), the system performs a partial memory allocation that suffices for the current decoding task without over-provisioning memory resources.
3Reliability
If the decoder uses the maximum number of reference frames for HD decoding, then decoding accuracy is maintained, but the memory requirement of 24.9 Mbytes is not feasible for many systems
Solution Approach 1:
The system dynamically adjusts memory allocation based on the actual decoding requirements of each GOP structure. By analyzing the GOP beforehand and determining the actual number M of reference frames needed, the system maintains sufficient memory for accurate decoding while avoiding excessive allocation, thus preserving reliability without over-consuming resources.
Data Source
AI summary
This invention is a method of memory saving in compressed video decoding. For each group of pictures the method determines whether less than the maximum number N reference frames are needed for decoding. If so, then a memory is configuring for the needed number M reference frames less than N. If not, the memory is configured for N reference frames. The group of pictures is decoding the group of pictures using the configured number of reference frames.