This is a page that curates AI-related papers published worldwide. All content here is summarized using Google Gemini and operated on a non-profit basis. Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.
This paper presents a novel method for integrated image restoration, an important task in low-level vision. Existing methods are either task-specific and have limited generalization ability to various types of corruptions, or they suffer from closed-set constraints as they are trained on paired datasets. To address this, we propose a dataset-free integrated approach using recursive posterior sampling with pre-trained latent diffusion models. The method integrates multimodal understanding models to provide semantic prior information for generative models under task-independent conditions, uses lightweight modules to align corrupted inputs with the generative preferences of the diffusion model, and utilizes recursive refinement for posterior sampling. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art methods, validating its effectiveness and robustness. Code and data will be made available at https://github.com/AMAP-ML/LD-RPS .
