Real-time Semantic Segmentation with Context Aggregation Network

TL;DR

This paper introduces the Context Aggregation Network, a dual-branch convolutional neural network designed for real-time semantic segmentation that balances high accuracy with low computational costs, suitable for autonomous systems.

Contribution

The paper presents a novel dual-branch architecture with a lightweight high-resolution branch and an efficient context branch, achieving state-of-the-art accuracy with significantly reduced computation.

Findings

Achieves 75.9% mIOU on Cityscapes at 76 FPS

Attains 63.5% mIOU on UAVid dataset at 15 FPS

Outperforms existing methods in speed and accuracy trade-offs

Abstract

With the increasing demand of autonomous systems, pixelwise semantic segmentation for visual scene understanding needs to be not only accurate but also efficient for potential real-time applications. In this paper, we propose Context Aggregation Network, a dual branch convolutional neural network, with significantly lower computational costs as compared to the state-of-the-art, while maintaining a competitive prediction accuracy. Building upon the existing dual branch architectures for high-speed semantic segmentation, we design a cheap high resolution branch for effective spatial detailing and a context branch with light-weight versions of global aggregation and local distribution blocks, potent to capture both long-range and local contextual dependencies required for accurate semantic segmentation, with low computational overheads. We evaluate our method on two semantic segmentation datasets, namely Cityscapes dataset and UAVid dataset. For Cityscapes test set, our model achieves state-of-the-art results with mIOU of 75.9%, at 76 FPS on an NVIDIA RTX 2080Ti and 8 FPS on a Jetson Xavier NX. With regards to UAVid dataset, our proposed network achieves mIOU score of 63.5% with high execution speed (15 FPS).