# Scan Clusters, Not Pixels: A Cluster-Centric Paradigm for Efficient Ultra-high-definition Image Restoration

arXiv: https://arxiv.org/abs/2602.21917 · HTML: https://mlsanity.com/paper/2602.21917/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 49 references that are in the map.
- First posted on arXiv 25 Feb 2026.

## Builds on (most cited first, 30 of 32)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [VMamba: Visual State Space Model](https://mlsanity.com/paper/2401.10166/) (2024)
- [Uformer: A General U-Shaped Transformer for Image Restoration](https://mlsanity.com/paper/2106.03106/) (2021)
- [MetaFormer Is Actually What You Need for Vision](https://mlsanity.com/paper/2111.11418/) (2021)
- [MambaIR: A Simple Baseline for Image Restoration with State-Space Model](https://mlsanity.com/paper/2402.15648/) (2024)
- [Rethinking Coarse-to-Fine Approach in Single Image Deblurring](https://mlsanity.com/paper/2108.05054/) (2021)
- [Vision Transformers for Single Image Dehazing](https://mlsanity.com/paper/2204.03883/) (2022)
- [Learning A Sparse Transformer Network for Effective Image Deraining](https://mlsanity.com/paper/2303.11950/) (2023)
- [Learning to Enhance Low-Light Image via Zero-Reference Deep Curve Estimation](https://mlsanity.com/paper/2103.00860/) (2021)
- [Ultra-High-Definition Low-Light Image Enhancement: A Benchmark and Transformer-Based Method](https://mlsanity.com/paper/2212.11548/) (2022)
- [Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring](https://mlsanity.com/paper/2211.12250/) (2022)
- [EfficientVMamba: Atrous Selective Scan for Light Weight Visual Mamba](https://mlsanity.com/paper/2403.09977/) (2024)
- [Stripformer: Strip Transformer for Fast Image Deblurring](https://mlsanity.com/paper/2204.04627/) (2022)
- [A Model-driven Deep Neural Network for Single Image Rain Removal](https://mlsanity.com/paper/2005.01333/) (2020)
- [MambaIRv2: Attentive State Space Restoration](https://mlsanity.com/paper/2411.15269/) (2024)
- [Embedding Fourier for Ultra-High-Definition Low-Light Image Enhancement](https://mlsanity.com/paper/2302.11831/) (2023)
- [MB-TaylorFormer: Multi-branch Efficient Transformer Expanded by Taylor Formula for Image Dehazing](https://mlsanity.com/paper/2308.14036/) (2023)
- [Structure-Preserving Deraining with Residue Channel Prior Guidance](https://mlsanity.com/paper/2108.09079/) (2021)
- [FreqMamba: Viewing Mamba from a Frequency Perspective for Image Deraining](https://mlsanity.com/paper/2404.09476/) (2024)
- [Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining](https://mlsanity.com/paper/2404.01547/) (2024)
- [Wave-Mamba: Wavelet State Space Model for Ultra-High-Definition Low-Light Image Enhancement](https://mlsanity.com/paper/2408.01276/) (2024)
- [Sparse Sampling Transformer with Uncertainty-Driven Ranking for Unified Removal of Raindrops and Rain Streaks](https://mlsanity.com/paper/2308.14153/) (2023)
- [MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration](https://mlsanity.com/paper/2412.20066/) (2024)
- [Correlation Matching Transformation Transformers for UHD Image Restoration](https://mlsanity.com/paper/2406.00629/) (2024)
- [FourierMamba: Fourier Learning Integration with State Space Models for Image Deraining](https://mlsanity.com/paper/2405.19450/) (2024)
- [Efficient Visual State Space Model for Image Deblurring](https://mlsanity.com/paper/2405.14343/) (2024)
- [From Zero to Detail: Deconstructing Ultra-High-Definition Image Restoration from Progressive Spectral Perspective](https://mlsanity.com/paper/2503.13165/) (2025)

## Built on by (most cited first, 8 of 8)

- [SP-MoMamba: Superpixel-driven Mixture of State Space Experts for Efficient Image Super-Resolution](https://mlsanity.com/paper/2605.25892/) (2026)
- [MambaRaw: Selective State Space Modeling for Efficient 4K Raw Image Reconstruction](https://mlsanity.com/paper/2606.24479/) (2026)
- [Backbone-Agnostic Stochastic Perturbation Learning for End-to-End Real-World Image Dehazing](https://mlsanity.com/paper/2607.11623/) (2026)
- [CoDe-SSM: Context-Detail Decoupled State Space Model for Efficient UHD Image Restoration](https://mlsanity.com/paper/2607.29595/) (2026)
- [Token Clustering and Semantic Sequence Mamba for Hyperspectral Image Classification](https://mlsanity.com/paper/2609.28580/) (2026)
- [PhoenixSR: Generative Heterogeneous Distillation Unleashes Efficient Models for Real-World Super-Resolution](https://mlsanity.com/paper/2609.30988/) (2026)
- [TaskIR: Task-Driven Image Restoration via Degradation Adaptation and Task Feedback](https://mlsanity.com/paper/2609.31170/) (2026)
- [Pre-training, Reasoning, Benchmarking: X-ray Report Generation on CheXpert Plus Dataset](https://mlsanity.com/paper/2610.08813/) (2026)

## Abstract (arXiv)

Ultra-High-Definition (UHD) image restoration is trapped in a scalability crisis: existing models, bound to pixel-wise operations, demand unsustainable computation. While state space models (SSMs) like Mamba promise linear complexity, their pixel-serial scanning remains a fundamental bottleneck for the millions of pixels in UHD content. We ask: must we process every pixel to understand the image? This paper introduces C2SSM, a visual state space model that breaks this taboo by shifting from pixel-serial to cluster-serial scanning. Our core discovery is that the rich feature distribution of a UHD image can be distilled into a sparse set of semantic centroids via a neural-parameterized mixture model. C2SSM leverages this to reformulate global modeling into a novel dual-path process: it scans and reasons over a handful of cluster centers, then diffuses the global context back to all pixels through a principled similarity distribution, all while a lightweight modulator preserves fine details. This cluster-centric paradigm achieves a decisive leap in efficiency, slashing computational costs while establishing new state-of-the-art results across five UHD restoration tasks. More than a solution, C2SSM charts a new course for efficient large-scale vision: scan clusters, not pixels.