Rethinking BiSeNet For Real-time Semantic Segmentation

TL;DR

This paper introduces the STDC network, a novel efficient architecture for real-time semantic segmentation that reduces redundancy and improves speed and accuracy on Cityscapes and CamVid datasets.

Contribution

The paper proposes the STDC network with a new feature aggregation method and a Detail Aggregation module, enhancing efficiency and segmentation performance.

Findings

Achieves 71.9% mIoU at 250.4 FPS on Cityscapes

Outperforms recent methods by 45.2% in speed

Maintains high accuracy at higher resolutions

Abstract

BiSeNet has been proved to be a popular two-stream network for real-time segmentation. However, its principle of adding an extra path to encode spatial information is time-consuming, and the backbones borrowed from pretrained tasks, e.g., image classification, may be inefficient for image segmentation due to the deficiency of task-specific design. To handle these problems, we propose a novel and efficient structure named Short-Term Dense Concatenate network (STDC network) by removing structure redundancy. Specifically, we gradually reduce the dimension of feature maps and use the aggregation of them for image representation, which forms the basic module of STDC network. In the decoder, we propose a Detail Aggregation module by integrating the learning of spatial information into low-level layers in single-stream manner. Finally, the low-level features and deep features are fused to predict the final segmentation results. Extensive experiments on Cityscapes and CamVid dataset demonstrate the effectiveness of our method by achieving promising trade-off between segmentation accuracy and inference speed. On Cityscapes, we achieve 71.9% mIoU on the test set with a speed of 250.4 FPS on NVIDIA GTX 1080Ti, which is 45.2% faster than the latest methods, and achieve 76.8% mIoU with 97.0 FPS while inferring on higher resolution images.