Integrating Amortized Inference with Diffusion Models for Learning Clean Distribution from Corrupted Images
원문 초록
Diffusion models (DMs) have emerged as powerful generative models for solving inverse problems, offering a good approximation of prior distributions of real-world image data. Typically, diffusion models rely on large-scale clean signals to accurately learn the score functions of ground truth clean image distributions. However, such a requirement for large amounts of clean data is often impractical in real-world applications, especially in fields where data samples are expensive to obtain. To address this limitation, in this work, we introduce \emph{FlowDiff}, a novel joint training paradigm that leverages a conditional normalizing flow model to facilitate the training of diffusion models on corrupted data sources. The conditional normalizing flow try to learn to recover clean images through a novel amortized inference mechanism, and can thus effectively facilitate the diffusion model's training with corrupted data. On the other side, diffusion models provide strong priors which in turn improve the quality of image recovery. The flow model and the diffusion model can therefore promote each other and demonstrate strong empirical performances. Our elaborate experiment shows that FlowDiff can effectively learn clean distributions across a wide range of corrupted data sources, such as noisy and blurry images. It consistently outperforms existing baselines with significant margins under identical conditions. Additionally, we also study the learned diffusion prior, observing its superior performance in downstream computational imaging tasks, including inpainting, denoising, and deblurring.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
EAMamba: Efficient All-Around Vision State Space Model for Image Restoration
Visual-Instructed Degradation Diffusion for All-in-One Image Restoration
Plug-and-Play Linear Attention for Pre-trained Image and Video Restoration Models
Multi-Step Guided Diffusion for Image Restoration on Edge Devices: Toward Lightweight Perception in Embodied AI
Plug-and-Play Posterior Sampling for Blind Inverse Problems
Joint Flow And Feature Refinement Using Attention For Video Restoration
토론
이 논문과 연결된 포럼 글이 아직 없습니다.