Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers
This work provides a new architectural paradigm for semantic segmentation, offering a powerful alternative to traditional FCNs for researchers and practitioners in computer vision.
This paper rethinks semantic segmentation as a sequence-to-sequence prediction task, employing a pure transformer to encode images as patch sequences without convolutions or resolution reduction. The resulting model, SETR, achieved new state-of-the-art performance on ADE20K with 50.28% mIoU and Pascal Context with 55.83% mIoU, also showing competitive results on Cityscapes.
Most recent semantic segmentation methods adopt a fully-convolutional network (FCN) with an encoder-decoder architecture. The encoder progressively reduces the spatial resolution and learns more abstract/semantic visual concepts with larger receptive fields. Since context modeling is critical for segmentation, the latest efforts have been focused on increasing the receptive field, through either dilated/atrous convolutions or inserting attention modules. However, the encoder-decoder based FCN architecture remains unchanged. In this paper, we aim to provide an alternative perspective by treating semantic segmentation as a sequence-to-sequence prediction task. Specifically, we deploy a pure transformer (ie, without convolution and resolution reduction) to encode an image as a sequence of patches. With the global context modeled in every layer of the transformer, this encoder can be combined with a simple decoder to provide a powerful segmentation model, termed SEgmentation TRansformer (SETR). Extensive experiments show that SETR achieves new state of the art on ADE20K (50.28% mIoU), Pascal Context (55.83% mIoU) and competitive results on Cityscapes. Particularly, we achieve the first position in the highly competitive ADE20K test server leaderboard on the day of submission.