QDM: Quadtree-Based Region-Adaptive Sparse Diffusion Models for Efficient Image Super-Resolution

2025-03-15

AI Summary

Existing deep learning-based super-resolution methods perform uniform pixel-wise computations across the entire image, leading to significant computational redundancy in homogeneous regions. To address this, we propose Quadtree-guided Adaptive Diffusion (QADiff), the first framework integrating quadtree-based spatial partitioning with a sparse diffusion mechanism to enable fine-grained, region-adaptive computation—activating high-resolution reconstruction exclusively in texture-rich regions. We further introduce a dual-stream mask-guided architecture that facilitates mask-driven feature interaction and region-adaptive sampling, enabling dynamic trade-offs between reconstruction quality and computational efficiency. Evaluated on medical imaging modalities such as CT, QADiff reduces FLOPs by up to 62% while matching or surpassing state-of-the-art methods in PSNR and SSIM. The method thus achieves high-fidelity reconstruction without compromising deployability on edge devices.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Deep learning-based super-resolution (SR) methods often perform pixel-wise computations uniformly across entire images, even in homogeneous regions where high-resolution refinement is redundant. We propose the Quadtree Diffusion Model (QDM), a region-adaptive diffusion framework that leverages a quadtree structure to selectively enhance detail-rich regions while reducing computations in homogeneous areas. By guiding the diffusion with a quadtree derived from the low-quality input, QDM identifies key regions-represented by leaf nodes-where fine detail is essential and applies minimal refinement elsewhere. This mask-guided, two-stream architecture adaptively balances quality and efficiency, producing high-fidelity outputs with low computational redundancy. Experiments demonstrate QDM's effectiveness in high-resolution SR tasks across diverse image types, particularly in medical imaging (e.g., CT scans), where large homogeneous regions are prevalent. Furthermore, QDM outperforms or is comparable to state-of-the-art SR methods on standard benchmarks while significantly reducing computational costs, highlighting its efficiency and suitability for resource-limited environments. Our code is available at https://github.com/linYDTHU/QDM.

Problem

• Efficient image super-resolution with reduced computational redundancy
• Selective enhancement of detail-rich regions using quadtree structure
• Improved performance in medical imaging with large homogeneous areas

Innovation

• Quadtree structure for selective region enhancement
• Mask-guided two-stream architecture for efficiency
• Reduced computational redundancy in homogeneous areas