A Review of various Architectures for Image Segmentation
Nagamani Gonthina, L V Narasimha Prasad · 2024
Identifying the objects of interest is one of main task in computer vision. To identify these objects in an image, we segment the image to homogeneous regions and extract the objects of interest. This entire process is termed as image segmentation. Various techniques like clustering, thresholding, watershed, neural networks, etc. are used to segment an image. To segment an image, it is to be fed as input to the segmentation architecture. We have various predefined architectures for the purpose of segmentation as well as various types of learning. In this study, we present various architectures used for image segmentation including types of learning in which they are implemented and discuss the pros and cons of each architecture.