arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.12533cs.CV
收藏

Visformer:视觉友好型Transformer

Visformer: The Vision-friendly Transformer

Beihang University(北京航空航天大学);Johns Hopkins University(约翰斯·霍普金斯大学);Hangzhou Innovation Institute of Beihang University(北京航空航天大学杭州创新研究院);Zhengzhou University(郑州大学);University of Science and Technology of China(中国科学技术大学);Xidian University(西安电子科技大学)

Zhengsu Chen, Lingxi Xie, Jianwei Niu, Xuefeng Liu, Longhui Wei, Qi Tian

摘要:

在过去的一年里,我们见证了Transformer模块在解决视力问题方面的快速发展。虽然一些研究人员已经证明了基于transformer的模型具有良好的拟合数据的能力,但仍然有越来越多的证据表明,这些模型遭受过拟合,特别是当训练数据有限时。本文提供了一个实证研究,通过执行一步一步的操作,逐步过渡到基于卷积的模型基于transformer的模型。我们在过渡过程中获得的结果为改善视觉识别提供了有用的信息。基于这些观察,我们提出了一个新的架构命名为Visformer,这是从“视觉友好的变压器”的缩写。在计算复杂性相同的情况下,Visformer在ImageNet分类精度方面优于基于transformer和基于卷积的模型,并且当模型复杂度较低或训练集较小时,优势变得更加明显。该代码可在https://github.com/danczs/Visformer上获得。

英文摘要:

The past year has witnessed the rapid development of applying the Transformer module to vision problems. While some researchers have demonstrated that Transformer-based models enjoy a favorable ability of fitting data, there are still growing number of evidences showing that these models suffer over-fitting especially when the training data is limited. This paper offers an empirical study by performing step-by-step operations to gradually transit a Transformer-based model to a convolution-based model. The results we obtain during the transition process deliver useful messages for improving visual recognition. Based on these observations, we propose a new architecture named Visformer, which is abbreviated from the `Vision-friendly Transformer'. With the same computational complexity, Visformer outperforms both the Transformer-based and convolution-based models in terms of ImageNet classification accuracy, and the advantage becomes more significant when the model complexity is lower or the training set is smaller. The code is available at https://github.com/danczs/Visformer.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑