Adaptive Frame Compression for Parallel Video and CV/NN Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Camera modules in consumer devices face challenges in efficiently utilizing limited embedded SRAM for parallel execution of multiple programs due to high noise reduction requirements, leading to insufficient memory for simultaneous video and computer vision/neural network processing.
Innovation Solution
An image compression system with a frame buffer, TNR and 2DNR circuits, and a controller that adaptively adjusts compression/decompression ratios to manage memory usage, allowing parallel execution of programs like video and CV/NN tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If temporal noise reduction and spatial noise reduction are implemented to handle sensor noises, then noise reduction effectiveness is improved, but memory space consumption increases
Solution Approach 1:
The patent implements dynamic adjustment of the compression ratio based on parallel execution requirements. The controller monitors whether parallel execution is needed and adaptively changes the compression ratio of reference frames stored in frame buffer, thereby dynamically allocating memory resources between noise reduction operations and parallel program execution.
Solution Approach 2:
The patent changes the compression ratio parameter of reference frames in the frame buffer based on system state. When parallel execution is required, the compression ratio is increased to reduce memory usage; when parallel execution is not required, the compression ratio is decreased to improve noise reduction quality. This parameter adjustment resolves the contradiction between memory consumption and noise reduction effectiveness.
2Quantity of substance
If compression ratio is increased to release memory for parallel programs, then memory availability for parallel execution is improved, but noise reduction quality deteriorates
Solution Approach 1:
The system dynamically switches between different compression ratios based on whether parallel execution is needed. The controller monitors the execution state and adjusts the compression ratio in real-time, making the system flexible enough to prioritize either memory availability or noise reduction quality depending on current requirements.
Solution Approach 2:
The patent implements periodic adjustment of compression ratio in response to parallel execution demands. The controller periodically evaluates whether parallel programs need to be executed and adjusts the compression ratio accordingly, creating a rhythmic pattern of memory allocation that balances both requirements over time.
3Productivity
If embedded SRAM is used for parallel program execution, then parallel processing capability is improved, but the complexity of memory management increases
Solution Approach 1:
The controller automatically monitors system state and adjusts compression ratios without external intervention. The system self-manages memory resources by detecting when parallel execution is needed and autonomously adjusting the compression ratio to free up necessary memory space, reducing the complexity of manual memory management.
Solution Approach 2:
The patent implements a feedback mechanism where the controller continuously monitors whether parallel execution is required and uses this feedback to adjust the compression ratio. This closed-loop control system automatically balances memory allocation between noise reduction and parallel processing, simplifying memory management through intelligent feedback-driven adjustment.
Data Source
AI summary
An image compression system includes a memory having a frame buffer, a temporal noise reduction (TNR) circuit, a compression circuit, and a controller. The temporal noise reduction circuit is configured to receive a frame from an image sensor and blend the frame with a decompressed reference frame from the frame buffer into a temporal noise reduction frame. The compression circuit is configured to receive the temporal noise reduction frame from the TNR circuit and compress the temporal noise reduction frame into a compressed reference frame utilizing a ratio in the frame buffer, wherein the compression circuit is further configured to decompress the compressed reference frame in the frame buffer by the ratio as the decompressed reference frame. The controller is configured to adaptively increase the ratio according to a number of program(s) to ensure an adequate memory space of the memory for executing the program(s).