Learning Multi-level Region Consistency with Dense Multi-label Networks for Semantic Segmentation

Learning Multi-level Region Consistency with Dense Multi-label Networks for Semantic Segmentation

Tong Shen, Guosheng Lin, Chunhua Shen, Ian Reid

Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence
Main track. Pages 2708-2714. https://doi.org/10.24963/ijcai.2017/377

Semantic image segmentation is a fundamental task in image understanding. Per-pixel semantic labelling of an image benefits greatly from the ability to consider region consistency both locally and globally. However, many Fully Convolutional Network based methods do not impose such consistency, which may give rise to noisy and implausible predictions. We address this issue by proposing a dense multi-label network module that is able to encourage the region consistency at different levels. This simple but effective module can be easily integrated into any semantic segmentation systems. With comprehensive experiments, we show that the dense multi-label can successfully remove the implausible labels and clear the confusion so as to boost the performance of semantic segmentation systems.
Keywords:
Machine Learning: Neural Networks
Machine Learning: Deep Learning