A Brief Survey on Weakly Supervised Semantic Segmentation

Semantic Segmentation is the process of assigning a label to every pixel in image that share same semantic properties and stays challenging task computer vision. In recent years, due large availability training data performance segmentation has been greatly improved by using deep learning techniques. A number novel methods have proposed. However, some crucial fields we can't assure sufficient learn model achieves high accuracy. This paper aims provide brief survey research efforts on deep-learning-based limited labeled focus our weakly-supervised methods. expected familiarize readers with progress challenges weakly supervised era present several valuable growing points this field.