OBJECT SEGMENTATION METHOD AND APPARATUS, COMPUTER READABLE STORAGE MEDIUM, AND COMPUTER DEVICE

摘要

An object segmentation method and apparatus, a computer readable storage medium, and a computer device, comprising: acquiring an initial feature map of an image to be processed (S102); inputting the initial feature map into a continuous top-down network and bottom-up network to calculate a first stage feature map, and respectively updating the first stage feature map in the top-down network and the bottom-up network on the basis of regional context encoding RCE to obtain an updated first stage feature map (S104); using the updated first stage feature map as an input to implement iterative fusion and update calculation until a preset number of iterations is reached to obtain a target feature map (S106); and, on the basis of the target feature map, implementing object segmentation on the image to be processed, the target feature map being a feature map obtained from the bottom-up network (S108).