OmniScaleSR: Unleashing Scale-Controlled Diffusion Prior for Faithful and Realistic Arbitrary-Scale Image Super-Resolution

2025-12-04

AI Summary

Existing arbitrary-scale super-resolution (ASSR) methods struggle to balance perceptual quality and fidelity: implicit neural representation (INR)-based approaches lack fine-detail synthesis capability, while diffusion-based Real-ISR, though leveraging strong priors, lacks explicit scale control—leading to hallucinations or blur at extreme upscaling factors. This paper proposes OmniScaleSR, the first framework integrating **native diffusion-compatible explicit scale control** with implicit scale adaptation, enabling content- and scale-aware diffusion modulation. A multi-domain fidelity enhancement module further improves reconstruction accuracy. Built upon a pre-trained diffusion prior, OmniScaleSR generalizes to arbitrary scaling factors without fine-tuning. Experiments demonstrate significant superiority over state-of-the-art methods on both bicubic and real-world degradation benchmarks, particularly excelling at large-scale factors by substantially enhancing detail realism and structural fidelity.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Arbitrary-scale super-resolution (ASSR) overcomes the limitation of traditional super-resolution (SR) methods that operate only at fixed scales (e.g., 4x), enabling a single model to handle arbitrary magnification. Most existing ASSR approaches rely on implicit neural representation (INR), but its regression-driven feature extraction and aggregation intrinsically limit the ability to synthesize fine details, leading to low realism. Recent diffusion-based realistic image super-resolution (Real-ISR) models leverage powerful pre-trained diffusion priors and show impressive results at the 4x setting. We observe that they can also achieve ASSR because the diffusion prior implicitly adapts to scale by encouraging high-realism generation. However, without explicit scale control, the diffusion process cannot be properly adjusted for different magnification levels, resulting in excessive hallucination or blurry outputs, especially under ultra-high scales. To address these issues, we propose OmniScaleSR, a diffusion-based realistic arbitrary-scale SR framework designed to achieve both high fidelity and high realism. We introduce explicit, diffusion-native scale control mechanisms that work synergistically with implicit scale adaptation, enabling scale-aware and content-aware modulation of the diffusion process. In addition, we incorporate multi-domain fidelity enhancement designs to further improve reconstruction accuracy. Extensive experiments on bicubic degradation benchmarks and real-world datasets show that OmniScaleSR surpasses state-of-the-art methods in both fidelity and perceptual realism, with particularly strong performance at large magnification factors. Code will be released at https://github.com/chaixinning/OmniScaleSR.

Problem

• Overcomes limitations of fixed-scale super-resolution methods
• Addresses excessive hallucination or blur in diffusion-based arbitrary-scale SR
• Enhances fidelity and realism in arbitrary-scale image super-resolution

Innovation

• Diffusion-based framework for arbitrary-scale super-resolution
• Explicit scale control mechanisms integrated with diffusion process
• Multi-domain fidelity enhancement designs for improved accuracy