SEMANTIC SEGMENTATION MODEL TRAINING METHOD, DEVICE, COMPUTER APPARATUS AND STORAGE MEDIUM

摘要

The present application provides a semantic segmentation model training method, device, computer apparatus and storage medium, the method comprising: building a training sample set; inputting the training sample set into a deep network model and performing training; further inputting into a weighting transfer function and performing training to obtain a boundary prediction mask parameter; and building a semantic segmentation model. The present application improves training efficiency.