Deep Learning Methods in Image Matting: A Survey

Image matting is a fundamental technique used to extract fine foreground image from given by estimating the opacity values of each pixel. It one key techniques in processing and has wide range applications practical scenarios, such as video editing. Deep learning demonstrated outstanding performance various tasks, making it popular research topic. In recent years, methods based on deep have gained significant attention due their superior performance. Therefore, this article presents comprehensive overview learning-based algorithms that been proposed years. This paper initially introduces frequently datasets production methods, along with basic principles traditional techniques. We then analyze detail introduce commonly evaluation metrics. Additionally, discusses application scenarios matting, conducts experiments illustrate limitations current outlines potential future directions field. Overall, can serve valuable reference for researchers are interested matting.