mlsanityData to 7 Oct 2026

MAMAT: 3D Mamba-Based Atmospheric Turbulence Removal and its Object Detection Capability

arXiv 2503.17700 ↗First posted 22 Mar 2025cs.CVcited by 5

On the map: State-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing

Abstract (arXiv)

Restoration and enhancement are essential for improving the quality of videos captured under atmospheric turbulence conditions, aiding visualization, object detection, classification, and tracking in surveillance systems. In this paper, we introduce a novel Mamba-based method, the 3D Mamba-Based Atmospheric Turbulence Removal (MAMAT), which employs a dual-module strategy to mitigate these distortions. The first module utilizes deformable 3D convolutions for non-rigid registration to minimize spatial shifts, while the second module enhances contrast and detail. Leveraging the advanced capabilities of the 3D Mamba architecture, experimental results demonstrate that MAMAT outperforms state-of-the-art learning-based methods, achieving up to a 3% improvement in visual quality and a 15% boost in object detection. It not only enhances visualization but also significantly improves object detection accuracy, bridging the gap between visual restoration and the effectiveness of surveillance applications.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Placed in Point clouds, detection and remote sensing, one of 10 directions in State-space sequence models (S4, Mamba). Builds on 27 of its 36 references that are in the map. First posted on arXiv 22 Mar 2025.

Builds on

27 of 36 references.

References · 27

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  4. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  5. 2020YOLOv4: Optimal Speed and Accuracy of Object Detectioncited by 1,624
  6. 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Modelcited by 1,051
  7. 2024YOLOv11: An Overview of the Key Architectural Enhancementscited by 414
  8. 2020VarifocalNet: An IoU-aware Dense Object Detectorcited by 230
  9. 2021TOOD: Task-aligned One-stage Object Detectioncited by 218
  10. 2021You Only Learn One Representation: Unified Network for Multiple Taskscited by 86
  11. 2021Accelerating Atmospheric Turbulence Simulation via Learned Phase-to-Space Transformcited by 50
  12. 2024nnMamba: 3D Biomedical Image Segmentation, Classification and Landmark Detection with State Space Modelcited by 45
15 more
  1. 2022Imaging through the Atmosphere using Turbulence Mitigation Transformercited by 31
  2. 2022Learning to restore images degraded by atmospheric turbulence using uncertaintycited by 23
  3. 2018Subsampled Turbulence Removal Networkcited by 22
  4. 2024Spatio-Temporal Turbulence Mitigation: A Translational Perspectivecited by 22
  5. 2022Atmospheric Turbulence Removal with Complex-Valued Convolutional Neural Networkcited by 21
  6. 2019ATFaceGAN: Single Face Image Restoration and Recognition from Atmospheric Turbulencecited by 18
  7. 2018Atmospheric turbulence mitigation for sequences with moving objects using recursive image fusioncited by 17
  8. 2019Atmospheric turbulence removal using convolutional neural networkcited by 12
  9. 2024Deep Learning Techniques for Atmospheric Turbulence Removal: A Reviewcited by 10
  10. 2024Turb-Seg-Res: A Segment-then-Restore Pipeline for Dynamic Videos with Atmospheric Turbulencecited by 7
  11. 2024DeTurb: Atmospheric Turbulence Mitigation with Deformable 3D Convolutions and 3D Swin Transformerscited by 6
  12. 2022Augmentation of Atmospheric Turbulence Effects on Thermal Adapted Object Detection Modelscited by 5
  13. 2022Object recognition in atmospheric turbulence scenescited by 4
  14. 2024Real-World Atmospheric Turbulence Correction via Domain Adaptationcited by 4
  15. 2024Detection of moving objects through turbulent media. Decomposition of Oscillatory vs Non-Oscillatory spatio-temporal vector fieldscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 2.

5 papers.

  1. 2025DMAT: An End-to-End Framework for Joint Atmospheric Turbulence Mitigation and Object Detectioncited by 1
  2. 2025RMFAT: Recurrent Multi-scale Feature Atmospheric Turbulence Mitigatorcited by 1
  3. 2025JDATT: A Joint Distillation Framework for Atmospheric Turbulence Mitigation and Target Detectioncited by 0
  4. 2026GSTurb: Gaussian Splatting for Atmospheric Turbulence Mitigationcited by 0
  5. 2026ReMATF: Recurrent Motion-Adaptive Multi-scale Turbulence Mitigation for Dynamic Scenescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024CFMW: Cross-modality Fusion Mamba for Robust Object Detection under Adverse Weather
  2. 2026Fore-Mamba3D: Mamba-based Foreground-Enhanced Encoding for 3D Object Detection
  3. 2026BVI-Mamba: Video Enhancement Using a Visual State-Space Model for Low-Light and Underwater Environments
  4. 2026SPLG-Mamba: Structure-Preserving Local-Global Mamba Network for Salient Object Detection in Optical Remote Sensing Images
  5. 2025MambaFusion: Height-Fidelity Dense Global Fusion for Multi-modal 3D Object Detection
  6. 2025M^3Detection: Multi-Frame Multi-Level Feature Fusion for Multi-Modal 3D Object Detection with Camera and 4D Imaging Radar
  7. 2025Diffusion-Based Restoration for Multi-Modal 3D Object Detection in Adverse Weather
  8. 2026YOLO12-MambaScan: An Efficient Object Detector with High-Frequency Enhancement and State-Space Modeling