BiT-MoFE: Unified illumination correction in capsule endoscopy

Abstract

As a prevalent non-invasive screening technique, Wireless Capsule Endoscopy is often hindered by poor image quality, including under-/overexposure and low light condition. While illumination correction based on diffusion modeling or frequency-domain decomposition has shown effectiveness, existing methods often (1) underexploit structural information, and (2) lack adaptive strategies for varying illumination degradations, leading to suboptimal restoration and unnecessary computation. To this end, we propose Brownian Bridge Diffusion Transformer-Mixture-of-Experts (BiT-MoFE), a unified adaptive framework that integrates the merits of the two paradigms for endoscopic illumination correction. We adopt a Brownian Bridge Diffusion framework, in which an efficient Transformer serves as the backbone network, and design a frequency-decomposed MoFEs module to explicitly handle illumination and image structure simultaneously. By dynamically selecting the most suitable experts conditioned on exposure cues and diffusion timesteps, our framework achieves a strong balance between restoration fidelity and computational efficiency. Extensive experiments on multiple public datasets demonstrate that BiT-MoFE achieves state-of-the-art performance on both exposure correction and low-light enhancement tasks.

Keywords

Brownian Bridge Diffusion Model / frequency domain decomposition / mixture of experts / unified endoscopic illumination correction

References

[1] Iddan G, Meron G, Glukhovsky A, Swain P. Wireless capsule endoscopy. Nature. 2000;405(6785):417. https://doi.org/10.1038/35013140

[2] Zhang Y, Bai L, Liu L, Ren H, Meng MQH. Deep reinforcement learning-based control for stomach coverage scanning of wireless capsule endoscopy. In: 2022 IEEE International Conference on Robotics and Biomimetics (ROBIO). IEEE; 2022:1-6. https://doi.org/10.1109/ROBIO55434.2022.10012018

[3] Bai L, Chen T, Wu Y, Wang A, Islam M, Ren H. LLCaps: learning to illuminate low-light capsule endoscopy with curved wavelet attention and reverse diffusion. In: Medical Image Computing and Computer Assisted Intervention – MICCAI 2023. Vol. 14229. Springer Nature; 2023:34-44. https://doi.org/10.1007/978-3-031-43999-5_4

[4] Bai L, Chen T, Tan Q, et al. EndoUIC: promptable diffusion transformer for unified illumination correction in capsule endoscopy. In: Medical Image Computing and Computer Assisted Intervention – MICCAI 2024. Vol. 15007. Springer Nature; 2024:296-306. https://doi.org/10.1007/978-3-031-72104-5_29

[5] Long M, Li Z, Xie X, Li G, Wang Z. Adaptive image enhancement based on guide image and fraction-power transformation for wireless capsule endoscopy. IEEE Trans Biomed Circuits Syst. 2018;12(5):993-1003. https://doi.org/10.1109/TBCAS.2018.2869530

[6] García-Vega A, Espinosa R, Ochoa-Ruiz G, et al. A novel hybrid endoscopic dataset for evaluating machine learning-based photometric image enhancement models. In: Advances in Computational Intelligence. Vol. 13612. Springer Nature; 2022:267-281. https://doi.org/10.1007/978-3-031-19493-1_22

[7] Chen T, Lyu Q, Bai L, et al. LighTDiff: surgical endoscopic image low-light enhancement with T-diffusion. In: Medical Image Computing and Computer Assisted Intervention – MICCAI 2024. Lecture Notes in Computer Science. Vol. 15006. Springer Nature; 2024:369-379. https://doi.org/10.1007/978-3-031-72089-5_35

[8] García-Vega A, Espinosa R, Ramírez-Guzmán L, et al. Multi-scale structural-aware exposure correction for endoscopic imaging. In: 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI). IEEE; 2023:1-5. https://doi.org/10.1109/ISBI53787.2023.10230724

[9] Li B, Xue K, Liu B, Lai YK. BBDM: image-to-image translation with Brownian bridge diffusion models. In: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2023:1952-1961. https://doi.org/10.1109/CVPR52729.2023.00194

[10] Rombach R, Blattmann A, Lorenz D, Esser P, Ommer B. High-resolution image synthesis with latent diffusion models. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2022:10684-10695. https://doi.org/10.1109/CVPR52688.2022.01042

[11] Wu Z, Wang H, Shi Y, Huang D, Zheng Y. A prior-driven lightweight network for endoscopic exposure correction. In: Medical Image Computing and Computer Assisted Intervention – MICCAI 2025. Vol. 15970. Springer Nature; 2026:13-23. https://doi.org/10.1007/978-3-032-05141-7_2

[12] Rukundo O, Pedersen M, Hovde Ø. Advanced image enhancement method for distant vessels and structures in capsule endoscopy. Comput Math Methods Med. 2017;2017:1-13. https://doi.org/10.1155/2017/9813165

[13] Afifi M, Derpanis KG, Ommer B, Brown MS. Learning multi-scale photo exposure correction. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2021:9153-9163. https://doi.org/10.1109/CVPR46437.2021.00904

[14] Zhang J, Zheng Z, Zhang J, Ren W. ECSNN: spiking neural networks for efficient exposure correction in endoscopy imaging. In: ICASSP 2025 – 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE; 2025:1-5. https://doi.org/10.1109/ICASSP49660.2025.10888020

[15] Huang J, Liu Y, Zhao F, et al. Deep Fourier-based exposure correction network with spatial-frequency interaction. In: Computer Vision – ECCV 2022. Vol. 13679. Springer Nature; 2022:163-180. https://doi.org/10.1007/978-3-031-19800-7_10

[16] Choo K, Jun Y, Yun M, Hwang SJ. Slice-consistent 3D volumetric brain CT-to-MRI translation with 2D Brownian bridge diffusion model. In: Medical Image Computing and Computer Assisted Intervention – MICCAI 2024. Vol. 15007. Springer Nature; 2024:657-667. https://doi.org/10.1007/978-3-031-72104-5_63

[17] Zheng X, Wen J, Zhuang J, et al. Diffusion-based virtual staining from polarimetric Mueller matrix imaging. In: Medical Image Computing and Computer Assisted Intervention – MICCAI 2025. Vol. 15960. Springer Nature; 2026:163-173. https://doi.org/10.1007/978-3-032-04927-8_16

[18] Bongratz F, Li Y, Elbaroudy S, Wachinger C. 3D shape-to-image Brownian bridge diffusion for brain MRI synthesis from cortical surfaces. In: Information Processing in Medical Imaging. Vol. 15829. Springer Nature; 2026:187-202. https://doi.org/10.1007/978-3-031-96628-6_13

[19] Esser P, Rombach R, Ommer B. Taming transformers for high-resolution image synthesis. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2021:12868-12878. https://doi.org/10.1109/CVPR46437.2021.01268

[20] Zamir SW, Arora A, Khan S, Hayat M, Khan FS, Yang MH. Restormer: efficient transformer for high-resolution image restoration. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2022:5718-5729. https://doi.org/10.1109/CVPR52688.2022.00564

[21] Peebles W, Xie S. Scalable diffusion models with transformers. In: 2023 IEEE/CVF International Conference on Computer Vision (ICCV). IEEE; 2023:4172-4182. https://doi.org/10.1109/ICCV51070.2023.00387

[22] Özdenizci O, Legenstein R. Restoring vision in adverse weather conditions with patch-based denoising diffusion models. IEEE Trans Pattern Anal Mach Intell. 2023;45(8):10346-10357. https://doi.org/10.1109/TPAMI.2023.3238179

[23] Perez E, Strub F, De Vries H, Dumoulin V, Courville A. FiLM: visual reasoning with a general conditioning layer. Proc AAAI Conf Artif Intell. 2018;32(1):1-10. https://doi.org/10.1609/aaai.v32i1.11671

[24] Zamfir E, Wu Z, Mehta N, et al. Complexity experts are task-discriminative learners for any image restoration. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2025:12753-12763. https://doi.org/10.1109/CVPR52734.2025.01190

[25] Song J, Meng C, Ermon S. Denoising diffusion implicit models, arXiv [Preprint]. 2020. https://doi.org/10.48550/ARXIV.2010.02502

[26] Smedsrud PH, Thambawita V, Hicks SA, et al. Kvasir-Capsule, a video capsule endoscopy dataset. Sci Data. 2021;8(1):142. https://doi.org/10.1038/s41597-021-00920-z

[27] Coelho P, Pereira A, Leite A, Salgado M, Cunha A. A deep learning approach for red lesions detection in video capsule endoscopies. In: Image Analysis and Recognition. Vol. 10882. Springer International Publishing; 2018:553-561. https://doi.org/10.1007/978-3-319-93000-8_63

[28] Potlapalli V, Zamir SW, Khan SH, Shahbaz Khan F. PromptIR: prompting for all-in-one image restoration. In: Advances in Neural Information Processing Systems. Vol. 36. Curran Associates, Inc.; 2023:71275-71293. https://doi.org/10.48550/arXiv.2306.13090

[29] Jiang H, Luo A, Fan H, Han S, Liu S. Low-light image enhancement with wavelet-based diffusion models. ACM Trans Graph. 2023;42(6):1-14. https://doi.org/10.1145/3618373

[30] Huang J, Liu Y, Fu X, et al. Exposure normalization and compensation for multiple-exposure correction. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2022:6033-6042. https://doi.org/10.1109/CVPR52688.2022.00595

[31] Zamir SW, Arora A, Khan S, et al. Learning enriched features for fast image restoration and enhancement. IEEE Trans Pattern Anal Mach Intell. 2023;45(2):1934-1948. https://doi.org/10.1109/TPAMI.2022.3167175

[32] Baek JH, Kim D, Choi SM, Lee HJ, Kim H, Koh YJ. Luminance-aware color transform for multiple exposure correction. In: 2023 IEEE/CVF International Conference on Computer Vision (ICCV). IEEE; 2023:6133-6142. https://doi.org/10.1109/ICCV51070.2023.00566

[33] Yang KF, Cheng C, Zhao SX, Zhang XS, Li YJ. Learning to adapt to light. Int J Comput Vis. 2023;131(4):1022-1041. https://doi.org/10.1007/s11263-022-01745-y

[34] Zhou D, Yang Z, Yang Y. Pyramid diffusion models for low-light image enhancement. In: Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence. International Joint Conferences on Artificial Intelligence Organization; 2023:1795-1803. https://doi.org/10.24963/ijcai.2023/199

[35] Xue M, He J, Wang W, Zhou M. Low-light image enhancement via clip-Fourier guided wavelet diffusion. ACM Trans Multimed Comput Commun Appl. 2025;21(11):1-22. https://doi.org/10.1145/3764933

[36] Yin Y, Xu D, Tan C, Liu P, Zhao Y, Wei Y. CLE diffusion: controllable light enhancement diffusion model. In: Proceedings of the 31st ACM International Conference on Multimedia. ACM; 2023:8145-8156. https://doi.org/10.1145/3581783.3612145

RIGHTS & PERMISSIONS

2026 The Author(s). Journal of Intelligent Medicine published by John Wiley & Sons Australia, Ltd on behalf of Tianjin University.