Linear Convergence of FISTA Iterates for Image Reconstruction Using Denoiser-Driven Regularization with Linear Denoisers
This paper proves that the iterates of PnP-FISTA and RED-APG, two image reconstruction algorithms using denoiser-driven regularization, converge linearly to a unique solution when applied to linear inverse problems and using a class of linear denoisers.
LoFi：基於隱式神經表示的可擴展局部圖像重建
LoFi 是一種基於坐標的局部圖像重建框架，它利用隱式神經表示，僅使用局部信息即可有效地重建圖像，並在內存使用和泛化能力方面具有顯著優勢。
패치 기반 확산 모델을 통해 학습된 이미지 사전 지식을 활용한 역문제 해결
이미지의 전체가 아닌 패치를 활용하여 학습된 확산 모델을 통해 효율적인 이미지 사전 지식을 학습하고, 이를 기반으로 다양한 역문제를 해결하는 방법론을 제시한다.
Multi-Scale Glance: A Novel Non-Semantic Context Descriptor for Improved Image Reconstruction
This paper introduces MS-Glance, a new image descriptor inspired by human perception, that leverages non-semantic context to enhance the quality of image reconstruction in tasks like implicit neural representation fitting and undersampled MRI reconstruction.
FINOLA: A Novel Method for Image Reconstruction and Self-Supervised Learning Using First-Order Norm+Linear Autoregression and One-Way Wave Equations
This paper introduces FINOLA, a novel image representation method based on the discovery that images share a set of one-way wave equations in a latent space, with each image corresponding to a unique solution generated from a learned initial condition.
ICNN을 사용한 변형 이미지 복원을 위한 원시-듀얼 알고리즘
본 논문에서는 입력-볼록 신경망(ICNN)으로 매개변수화된 정규화기를 사용하여 변형 이미지 복원 문제를 해결하기 위한 효율적인 원시-듀얼 알고리즘을 제안합니다.
BAGS: Blur Agnostic Gaussian Splatting for Image Reconstruction
BAGS introduces Blur Agnostic Gaussian Splatting to address image blur and improve scene reconstruction quality.
Stochastic ADMM Algorithm for Large-Scale Ptychography with Weighted Total Variation
Introducing a novel stochastic ADMM algorithm for large-scale ptychography with weighted total variation to enhance image reconstruction quality.
© 2024 by Linnk AI