Image Processing Decoupling for Lower Memory and Bandwidth Demand
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods require excessive memory usage and high bandwidth demands, leading to transfer delays and failures when processing multiple images, particularly due to the need for encoding and decoding at the client end.
Innovation Solution
Decouple the image processing by determining retrieval information and text descriptive information at a rendering end, then concatenating these to form a target parameter sent to a serving end for efficient image processing, thereby reducing memory and bandwidth pressure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If image encoding and decoding is performed at the client end, then image processing can be completed, but memory usage becomes excessive and bandwidth demands increase
Solution Approach 1:
The patent extracts the encoding and decoding operations from the client end and relocates them to the serving end. The client only needs to send original images and instruction text, while the serving end performs the resource-intensive encoding/decoding operations, thus reducing client memory usage and bandwidth demands.
Solution Approach 2:
The patent introduces an intermediary mechanism where instruction text (retrieval information) acts as a mediator between the client's processing needs and the serving end's execution. The client sends textual instructions rather than processed image data, and the serving end uses these instructions to retrieve and process the actual image data, reducing direct data transfer requirements.
2Productivity
If multiple images are processed simultaneously, then processing throughput increases, but transfer delays and failures increase due to high bandwidth demands
Solution Approach 1:
The patent segments the image processing workflow into distinct phases: the client segment sends lightweight instruction text and original images, while the serving end segment handles the heavy lifting of encoding, decoding, and processing. This segmentation allows multiple clients to send instructions simultaneously without competing for bandwidth during the intensive processing phase.
Solution Approach 2:
The patent implements preliminary action by having the client prepare and send instruction text in advance, which the serving end can then execute without waiting for additional data. The retrieval information is prepared beforehand and stored on the serving end, enabling rapid processing when multiple images need to be handled.
3Ease of operation
If the client performs all processing operations, then processing can be completed, but the system becomes complex and resource-intensive at the client end
Solution Approach 1:
The patent enables the serving end to perform self-service by automatically retrieving images based on instruction text and executing processing operations without requiring client intervention. The system uses automated workflows where the serving end manages its own resource allocation, encoding/decoding operations, and image retrieval based on received instructions.
Data Source
AI summary
Embodiments of the present disclosure provide an image processing method, an apparatus, an electronic device, and a storage medium. The method comprises: determining retrieval information corresponding to at least one image to be used; obtaining a target parameter to be sent to a first serving end by processing text descriptive information of the image to be used and the retrieval information, to cause the first serving end to perform image processing based on the target parameter.