Timezone: »

Glance-and-Gaze Vision Transformer
Qihang Yu · Yingda Xia · Yutong Bai · Yongyi Lu · Alan Yuille · Wei Shen

Thu Dec 09 08:30 AM -- 10:00 AM (PST) @ None #None

Recently, there emerges a series of vision Transformers, which show superior performance with a more compact model size than conventional convolutional neural networks, thanks to the strong ability of Transformers to model long-range dependencies. However, the advantages of vision Transformers also come with a price: Self-attention, the core part of Transformer, has a quadratic complexity to the input sequence length. This leads to a dramatic increase of computation and memory cost with the increase of sequence length, thus introducing difficulties when applying Transformers to the vision tasks that require dense predictions based on high-resolution feature maps.In this paper, we propose a new vision Transformer, named Glance-and-Gaze Transformer (GG-Transformer), to address the aforementioned issues. It is motivated by the Glance and Gaze behavior of human beings when recognizing objects in natural scenes, with the ability to efficiently model both long-range dependencies and local context. In GG-Transformer, the Glance and Gaze behavior is realized by two parallel branches: The Glance branch is achieved by performing self-attention on the adaptively-dilated partitions of the input, which leads to a linear complexity while still enjoying a global receptive field; The Gaze branch is implemented by a simple depth-wise convolutional layer, which compensates local image context to the features obtained by the Glance mechanism. We empirically demonstrate our method achieves consistently superior performance over previous state-of-the-art Transformers on various vision tasks and benchmarks.

Author Information

Qihang Yu (Johns Hopkins University)
Yingda Xia (Johns Hopkins University)
Yutong Bai (Johns Hopkins University)
Yongyi Lu
Alan Yuille (Johns Hopkins University)
Wei Shen (Shanghai Jiao Tong University)

More from the Same Authors