The field of image restoration and denoising is rapidly advancing, with a focus on developing more robust and efficient methods. Recent research has explored the use of diffusion models, which have shown significant potential in blind face restoration and image denoising. These models are able to learn the underlying structure of images and remove noise and degradation without requiring prior knowledge of the degradation model. Another area of focus is on improving the accuracy and efficiency of image restoration methods, such as super-resolution and denoising, by leveraging 3D geometric priors and incorporating unsupervised learning techniques. Noteworthy papers include Towards Robust Time-of-Flight Depth Denoising with Confidence-Aware Diffusion Model, which proposes a novel ToF denoising approach, and KernelFusion: Assumption-Free Blind Super-Resolution via Patch Diffusion, which introduces a zero-shot diffusion-based method for blind super-resolution. Additionally, Invert2Restore: Zero-Shot Degradation-Blind Image Restoration presents a training-free method for zero-shot image restoration, and Diffusion Image Prior introduces a blind image restoration method based on pretrained diffusion models.