Cars Can't Fly up in the Sky: Improving Urban-Scene Segmentation via Height-driven Attention Networks
This work addresses the problem of improving segmentation accuracy for urban-scene images, which have unique characteristics often ignored by existing methods, though it is incremental as it builds on prior segmentation models.
The paper tackles urban-scene semantic segmentation by proposing a height-driven attention network (HANet) that selectively emphasizes features based on pixel vertical positions, achieving a new state-of-the-art performance on the Cityscapes benchmark with a large margin among ResNet-101 based models.
This paper exploits the intrinsic features of urban-scene images and proposes a general add-on module, called height-driven attention networks (HANet), for improving semantic segmentation for urban-scene images. It emphasizes informative features or classes selectively according to the vertical position of a pixel. The pixel-wise class distributions are significantly different from each other among horizontally segmented sections in the urban-scene images. Likewise, urban-scene images have their own distinct characteristics, but most semantic segmentation networks do not reflect such unique attributes in the architecture. The proposed network architecture incorporates the capability exploiting the attributes to handle the urban scene dataset effectively. We validate the consistent performance (mIoU) increase of various semantic segmentation models on two datasets when HANet is adopted. This extensive quantitative analysis demonstrates that adding our module to existing models is easy and cost-effective. Our method achieves a new state-of-the-art performance on the Cityscapes benchmark with a large margin among ResNet-101 based segmentation models. Also, we show that the proposed model is coherent with the facts observed in the urban scene by visualizing and interpreting the attention map. Our code and trained models are publicly available at https://github.com/shachoi/HANet