mlsanityData to 7 Oct 2026

CEM-TUDASR: Computationally efficient multi-modality transformer based unsupervised domain adaptive super-resolution approach

arXiv 2609.11201 ↗First posted 10 Sep 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Wireless Capsule Endoscopy (WCE) enables non-invasive visualization of the gastrointestinal tract, but its miniaturized optics, sensor limitations, and wireless transmission constraints result in low-resolution images with reduced visibility of diagnostically important structures. This paper proposes CEM-TUDASR, a computationally efficient unsupervised Transformer-based super-resolution framework for WCE image enhancement without paired low-resolution (LR) and high-resolution (HR) training data. A domain-adaptive degradation network synthesizes realistic WCE-like LR images from HR conventional endoscopy images, reducing the domain gap and enabling effective unpaired learning. The SR generator integrates Deep Attention Blocks (DABs) and a Fusion Attention Block (FAB) to capture long-range contextual dependencies and fine local structures while preserving perceptual and structural fidelity. The model is trained on a curated dataset derived from Kvasir Capsule and evaluated on KID and GIANA for cross-dataset generalization. No-reference quality metrics, including BRISQUE, PIQE, NIQE, and the domain-specific EndoQM, show that CEM-TUDASR consistently outperforms existing unsupervised SR methods. Qualitative results further demonstrate improved restoration of mucosal textures, vascular patterns, and clinically relevant anatomical details. Cross-domain experiments on retinal images additionally demonstrate the adaptability of the framework. With only 2.67 million parameters and 169.94 GFLOPs, CEM-TUDASR achieves high-quality reconstruction while maintaining computational efficiency, making it suitable for resource-constrained clinical and embedded endoscopic applications.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 24 of its 55 references that are in the map. First posted on arXiv 10 Sep 2026.

Builds on

24 of 55 references.

References · 24

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  4. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  5. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  6. 2018Image Super-Resolution Using Very Deep Residual Channel Attention Networkscited by 1,256
  7. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 902
  8. 2021Designing a Practical Degradation Model for Deep Blind Image Super-Resolutioncited by 524
  9. 2020Learning Texture Transformer Network for Image Super-Resolutioncited by 346
  10. 2015Deeply-Recursive Convolutional Network for Image Super-Resolutioncited by 317
  11. 2017"Zero-Shot" Super-Resolution using Deep Internal Learningcited by 285
  12. 2021Unsupervised Degradation Representation Learning for Blind Super-Resolutioncited by 177
12 more
  1. 2018Deep Learning for Single Image Super-Resolution: A Brief Reviewcited by 158
  2. 2018Efficient and Accurate MRI Super-Resolution using a Generative Adversarial Network and 3D Multi-Level Densely Connected Networkcited by 64
  3. 2019Unsupervised Learning for Real-World Super-Resolutioncited by 64
  4. 2021ConvNets vs. Transformers: Whose Visual Representations are More Transferable?cited by 35
  5. 2020Deep Generative Adversarial Residual Convolutional Networks for Real-World Super-Resolutioncited by 17
  6. 2020Unsupervised Real Image Super-Resolution via Generative Variational AutoEncodercited by 16
  7. 2020EndoL2H: Deep Super-Resolution for Capsule Endoscopycited by 12
  8. 2020Unsupervised MRI Super-Resolution Using Deep External Learning and Guided Residual Dense Network with Multimodal Image Priorscited by 4
  9. 2020Stereo Endoscopic Image Super-Resolution Using Disparity-Constrained Parallel Attentioncited by 2
  10. 2020CT Super Resolution via Zero Shot Learningcited by 2
  11. 2022Unsupervised Representation Learning for 3D MRI Super Resolution with Degradation Adaptationcited by 2
  12. 2023An Unsupervised Framework for Joint MRI Super Resolution and Gibbs Artifact Removalcited by 2

Built on by

Not cited yet.