RTFormer: Efficient Design for Real-Time Semantic Segmentation with Transformer
This work addresses the need for efficient transformer-based models in real-time semantic segmentation, offering a practical solution for applications like autonomous driving and robotics.
The authors tackled the problem of real-time semantic segmentation by proposing RTFormer, a dual-resolution transformer that achieves a better performance-efficiency trade-off than CNN-based models, achieving state-of-the-art results on Cityscapes, CamVid, and COCOStuff benchmarks.
Recently, transformer-based networks have shown impressive results in semantic segmentation. Yet for real-time semantic segmentation, pure CNN-based approaches still dominate in this field, due to the time-consuming computation mechanism of transformer. We propose RTFormer, an efficient dual-resolution transformer for real-time semantic segmenation, which achieves better trade-off between performance and efficiency than CNN-based models. To achieve high inference efficiency on GPU-like devices, our RTFormer leverages GPU-Friendly Attention with linear complexity and discards the multi-head mechanism. Besides, we find that cross-resolution attention is more efficient to gather global context information for high-resolution branch by spreading the high level knowledge learned from low-resolution branch. Extensive experiments on mainstream benchmarks demonstrate the effectiveness of our proposed RTFormer, it achieves state-of-the-art on Cityscapes, CamVid and COCOStuff, and shows promising results on ADE20K. Code is available at PaddleSeg: https://github.com/PaddlePaddle/PaddleSeg.