EUVIMEDEUVIMED entdecken
Crossref

M2Restore: Mixture-of-Experts-Based Mamba-CNN Fusion Framework for All-in-One Image Restoration

Yongzhen Wang, Yongjun Li, Zhuoran Zheng, Xiao-Ping Zhang, Mingqiang Wei

IEEE Transactions on Image Processing · 2025 · 34 · 8086-8100

DOI: 10.1109/tip.2025.3638662 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2025. Verfasst von Yongzhen Wang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 76 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Natural images are often degraded by complex, composite degradations such as rain, snow, and haze, which adversely impact downstream vision applications. While existing image restoration efforts have achieved notable success, they are still hindered by two critical challenges: limited generalization across dynamically varying degradation scenarios and a suboptimal balance between preserving local details and modeling global dependencies. To overcome these challenges, we propose M2Restore, a novel Mixture-of-Experts (MoE)-based Mamba-CNN fusion framework for efficient and robust all-in-one image restoration. M2Restore introduces three key contributions: First, to boost the model's generalization across diverse degradation conditions, we exploit a CLIP-guided MoE gating mechanism that fuses task-conditioned prompts with CLIP-derived semantic priors. This mechanism is further refined via cross-modal feature calibration, which enables precise expert selection for various degradation types. Second, to jointly capture global contextual dependencies and fine-grained local details, we design a dual-stream architecture that integrates the localized representational strength of CNNs with the long-range modeling efficiency of Mamba. This integration enables collaborative optimization of global semantic relationships and local structural fidelity, preserving global coherence while enhancing detail restoration. Third, we introduce an edge-aware dynamic gating mechanism that adaptively balances global modeling and local enhancement by reallocating computational attention to degradation-sensitive regions. This targeted focus leads to more efficient and precise restoration. Extensive experiments across multiple image restoration benchmarks validate the superiority of M2Restore in both visual quality and quantitative performance. Code is available at https://github.com/yz-wang/M2Restore.

Zitieren

APA

Wang, Y., Li, Y., Zheng, Z., Zhang, X. P., & Wei, M. (2025). M2Restore: Mixture-of-Experts-Based Mamba-CNN Fusion Framework for All-in-One Image Restoration. IEEE Transactions on Image Processing, 34, 8086-8100. https://doi.org/10.1109/tip.2025.3638662

Vancouver

Wang Y, Li Y, Zheng Z, Zhang XP, Wei M. M2Restore: Mixture-of-Experts-Based Mamba-CNN Fusion Framework for All-in-One Image Restoration. IEEE Transactions on Image Processing. 2025;34:8086-8100. doi:10.1109/tip.2025.3638662

BibTeX
@article{Wang2025,
  author = {Wang, Yongzhen and Li, Yongjun and Zheng, Zhuoran and Zhang, Xiao-Ping and Wei, Mingqiang},
  title = {M2Restore: Mixture-of-Experts-Based Mamba-CNN Fusion Framework for All-in-One Image Restoration},
  journal = {IEEE Transactions on Image Processing},
  year = {2025},
  volume = {34},
  pages = {8086--8100},
  doi = {10.1109/tip.2025.3638662}
}
Kostenlos starten →