2D Image Compression Using Importance Scoring for 3D Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high cost and inefficiency of storing and processing large quantities of 2D images for 2D to 3D image reconstruction tasks, along with the challenge of maintaining image quality and accuracy in the reconstruction process, particularly for challenging scenes and invisible objects, are unresolved by existing methods.
Innovation Solution
A content-aware lossless compression method using a transformer network-based image compressor network to determine importance scores of images, retaining a selected subset and abandoning others, thereby reducing storage and computing resources while maintaining high reconstruction quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large quantity of 2D images are stored and processed for 2D to 3D image reconstruction tasks, then the reconstruction quality and accuracy are improved, but the storage cost and processing time increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-training an image compressor network to evaluate and select important images before the actual 3D reconstruction process. The network pre-assesses the contribution of each 2D image to the final reconstruction quality, allowing the system to select only the most valuable images in advance, thereby reducing processing time and computational resources while maintaining reconstruction accuracy.
2Measurement precision
If a large quantity of 2D images are stored and processed for 2D to 3D image reconstruction tasks, then the reconstruction quality and accuracy are improved, but the storage cost increases significantly
Solution Approach 1:
The patent applies the extraction principle by using the trained image compressor network to extract and identify only the essential and important 2D images from the complete set of available images. The network evaluates each image's contribution to reconstruction quality and extracts a minimal subset that contains all necessary information, thereby significantly reducing storage requirements while preserving reconstruction accuracy.
3Quantity of substance
If traditional compression methods are used to reduce storage resources, then storage efficiency is improved, but the image quality and accuracy deteriorate
Solution Approach 1:
The patent applies parameter changes by transforming the compression approach from uniform compression of all images to selective compression based on importance scores. The image compressor network assigns different importance parameters to different images based on their contribution to 3D reconstruction, allowing the system to compress or discard less important images while preserving high-quality versions of critical images, thus maintaining overall reconstruction quality while reducing storage resources.
Data Source
AI summary
The present disclosure relates to a method, a device, and a computer program product for compressing a two-dimensional image. The method includes determining a plurality of importance scores of a plurality of images by a trained image compressor network according to pixel values of the plurality of images. The method further includes selecting an image subset from the plurality of images according to the plurality of importance scores of the plurality of images. In addition, the method further includes compressing the plurality of images by retaining the selected image subset and abandoning the remaining images. In this way, high image reconstruction quality is maintained while a high compression ratio is achieved. Moreover, as a manual labeling or calibration process is avoided, a large-scale data set can be processed with less manual intervention and fewer computing resources.