Laplacian Pyramid Compression for Image Memory Bandwidth
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods require significant memory bandwidth for storing reduced images, which can be inefficient, especially when generating wide-angle images or 360° views using multiple camera images.
Innovation Solution
An image processing apparatus and method that generates a Laplacian pyramid from an image, compressing it using a compressor with an encoder, quantiser, and predictor to efficiently store the image data, optimizing memory usage by varying bit allocation based on value magnitude and type of data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reduced images are stored in memory for multiband blending, then the blending process can be performed, but memory bandwidth requirements become excessive
Solution Approach 1:
The patent extracts only the essential information from images by computing Laplacian pyramids, which represent images as a series of frames containing different frequency components. This extraction allows storing only the necessary data for blending rather than complete reduced images, significantly reducing memory bandwidth requirements while maintaining blending capability
Solution Approach 2:
The patent changes the representation parameters of image data by transforming images into Laplacian pyramid form with varying frequency components across different levels. This parameter transformation enables efficient compression and reduced memory bandwidth usage while preserving the information needed for multiband blending operations
2Quantity of substance
If compression is applied to reduce images before storage, then memory bandwidth is reduced, but compression operations become complicated
Solution Approach 1:
The patent segments the image into a Laplacian pyramid with multiple levels, where each level contains frames representing different frequency components. This segmentation allows for efficient compression at each level independently, simplifying the overall compression process while achieving significant memory bandwidth reduction
Solution Approach 2:
By transforming images into Laplacian pyramid representation, the patent changes the data structure parameters to a form that is inherently more compressible. This parameter change simplifies compression operations compared to compressing conventional reduced images, as the Laplacian structure naturally separates frequency information that can be efficiently encoded
Data Source
AI summary
An image processing apparatus is provided that comprises an input configured to receive an image and a Laplacian generator configured to generate, from the image, a Laplacian pyramid that represents the image as a series of frames that contain different frequency components of the image. The image processing apparatus also comprises a compressor configured to compress the Laplacian pyramid for writing to memory.