Learning a Discriminative Feature Network for Semantic Segmentation
This addresses accuracy issues in semantic segmentation for computer vision applications, representing a strong incremental improvement.
The paper tackles intra-class inconsistency and inter-class indistinction in semantic segmentation by proposing a Discriminative Feature Network (DFN) with Smooth and Border sub-networks, achieving state-of-the-art performance of 86.2% mean IOU on PASCAL VOC 2012 and 80.3% on Cityscapes.
Most existing methods of semantic segmentation still suffer from two aspects of challenges: intra-class inconsistency and inter-class indistinction. To tackle these two problems, we propose a Discriminative Feature Network (DFN), which contains two sub-networks: Smooth Network and Border Network. Specifically, to handle the intra-class inconsistency problem, we specially design a Smooth Network with Channel Attention Block and global average pooling to select the more discriminative features. Furthermore, we propose a Border Network to make the bilateral features of boundary distinguishable with deep semantic boundary supervision. Based on our proposed DFN, we achieve state-of-the-art performance 86.2% mean IOU on PASCAL VOC 2012 and 80.3% mean IOU on Cityscapes dataset.