New framework uses implicit neural representations for speckle denoising

Researchers have developed a novel training-free framework for denoising speckle noise in dynamic imaging. This method utilizes a spatiotemporal implicit neural representation combined with an aperture-aware maximum-likelihood formulation to reconstruct clear imagery from noisy observations. The approach explicitly models the spatial covariance of speckle, allowing it to adapt to different pupil geometries without retraining, and employs a matrix-free implementation for practical optimization. Results from simulations and laboratory experiments show enhanced spatial fidelity and temporal consistency compared to existing methods. AI

IMPACT This method could improve image quality in various applications by effectively removing speckle noise without requiring extensive training data.

RANK_REASON The cluster contains a research paper detailing a new technical approach to image processing. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• Implicit Neural Speckle Denoising
• Influence Flower
• ScienceCast

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →