Image Reconstruction Approaches Based on Fusion of Event Stream and Image Frame: A Survey

As computer vision rapidly advances, traditional frame-based image capture techniques encounter significant challenges in capturing high-speed dynamic scenes. Event cameras, as an emerging bio-inspired sensor, provide a novel solution for image reconstruction by outputting a stream of events associated with luminance changes. In this paper, we review the current research status and development trend of image reconstruction approaches based on event stream and image frame fusion. We first introduce the working principle of event cameras and their advantages in image reconstruction, and then discuss in detail the main approaches in the key subtasks of video frame interpolation, image quality enhancement, image super-resolution and image generation. Through comparative analyses, we summarize the performance of these approaches in processing dynamic scenes, improving image quality and enhancing detail, and discuss their potential and limitations in practical applications. The article is also summarized by discussing the main challenges currently faced and suggesting possible future directions.