Event-Diffusion: Event-Based Image Reconstruction and Restoration with Diffusion Models
Quanmin Liang,Xiawu Zheng,Kai Huang,Yan Zhang,Jie Chen,Yonghong Tian
DOI: https://doi.org/10.1145/3581783.3612462
2023-01-01
Abstract:Event cameras offer the advantages of low latency, high temporal resolution and HDR compared to conventional cameras. Due to the asynchronous and sparse nature of events, many existing algorithms cannot be directly applied, necessitating the reconstruction of intensity frames. However, existing reconstruction methods often result in artifacts and edge blurring due to noise and event accumulation. In this paper, we argue that the key to event-based image reconstruction is to enhance the edge information of objects and restore the artifacts in the reconstructed images. To explain, edge information is one of the most important features in the event stream, providing information on the shape and contour of objects. Considering the extraordinary capabilities of Denoising Diffusion Probabilistic Models (DDPMs) in image generation, reconstruction, and restoration, we propose a new framework which incorporate it into the reconstruction pipeline to obtain high-quality results which effectively remove artifacts and blur in reconstructed images. Specifically, we first extract edge information from the event stream using the proposed event-based denoising method. It employs the contrast maximization framework to remove noise from the event stream and extract clear object edge information. And then, the edge information is further adopted to our diffusion model, which is used to enhance the edges of objects in the reconstructed images, thus improving the restoration effect. Experimental results show that our method achieves significant improvements in the mean squared error (MSE), the structural similarity (SSIM), and the perceptual similarity (LPIPS) metrics, with average improvements of 40%, 15%, and 25%, respectively, compared to previous state-of-the-art models, and has good generalization performance.