COLI: A Hierarchical Efficient Compressor for Large Images

2025-07-15

AI Summary

To address the poor detail preservation and low efficiency in compressing high-resolution, large-field-of-view images, this paper proposes COLI, an efficient compression framework based on the Neural Representation Video (NeRV) model. Methodologically, COLI innovatively integrates a pretraining-fine-tuning paradigm with mixed-precision training to accelerate convergence, and introduces Hyper-Compression—a suite of techniques comprising weight hyper-compression, parallelized loss computation, and optimized coordinate embedding—within the implicit neural representation (INR) framework to significantly improve both compression ratio and reconstruction fidelity. Evaluated on two medical imaging datasets, COLI outperforms baseline methods in PSNR and SSIM while substantially reducing bits per pixel (bpp) and accelerating compression by 4×. To our knowledge, COLI is the first INR-based image compression method to achieve simultaneous optimization of high fidelity and high efficiency.

Technology Category

• Computer Vision: Representation Learning for Vision
• Machine Learning: Learning on the Edge & Model Compression
• Data Mining & Knowledge Management: Data Compression

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

The escalating adoption of high-resolution, large-field-of-view imagery amplifies the need for efficient compression methodologies. Conventional techniques frequently fail to preserve critical image details, while data-driven approaches exhibit limited generalizability. Implicit Neural Representations (INRs) present a promising alternative by learning continuous mappings from spatial coordinates to pixel intensities for individual images, thereby storing network weights rather than raw pixels and avoiding the generalization problem. However, INR-based compression of large images faces challenges including slow compression speed and suboptimal compression ratios. To address these limitations, we introduce COLI (Compressor for Large Images), a novel framework leveraging Neural Representations for Videos (NeRV). First, recognizing that INR-based compression constitutes a training process, we accelerate its convergence through a pretraining-finetuning paradigm, mixed-precision training, and reformulation of the sequential loss into a parallelizable objective. Second, capitalizing on INRs' transformation of image storage constraints into weight storage, we implement Hyper-Compression, a novel post-training technique to substantially enhance compression ratios while maintaining minimal output distortion. Evaluations across two medical imaging datasets demonstrate that COLI consistently achieves competitive or superior PSNR and SSIM metrics at significantly reduced bits per pixel (bpp), while accelerating NeRV training by up to 4 times.

Problem

• Efficient compression for high-resolution large images
• Preserving critical details in image compression
• Improving speed and ratio in INR-based compression

Innovation

• Accelerates convergence via pretraining and mixed-precision training
• Implements Hyper-Compression to enhance compression ratios
• Uses Neural Representations for Videos (NeRV) framework