arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2012.00364cs.CVcs.LG
收藏

预训练图像处理Transformer

Pre-Trained Image Processing Transformer

Peking University(北京大学);Huawei Technologies(华为技术有限公司);The University of Sydney(悉尼大学);Peng Cheng Laboratory(鹏城实验室)

Hanting Chen, Yunhe Wang, Tianyu Guo, Chang Xu, Yiping Deng, Zhenhua Liu, Siwei Ma, Chunjing Xu, Chao Xu, Wen Gao

摘要:

随着现代硬件的计算能力的强劲增长,预训练的深度学习模型(例如,BERT,GPT-3)在大规模数据集上学习的方法已经显示出其优于传统方法的有效性。这主要归功于Transformer及其变体体系结构的表示能力。在本文中,我们研究了低级计算机视觉任务(例如,去噪、超分辨率和去噪),并开发了一种新的预训练模型,即图像处理Transformer(IPT)。为了最大限度地挖掘Transformer的能力,我们提出利用著名的ImageNet基准来生成大量的损坏图像对。IPT模型在这些具有多头和多尾的图像上进行训练。此外,为了更好地适应不同的图像处理任务,还引入了对比学习。因此,经过微调后,预训练的模型可以有效地用于所需的任务。只有一个预训练模型,IPT在各种低级基准测试中的表现优于当前最先进的方法。代码可在https://github.com/huawei-noah/Pretrained-IPT和https://gitee.com/mindspore/mindspore/tree/master/model_zoo/research/cv/IPT上获得

英文摘要:

As the computing power of modern hardware is increasing strongly, pre-trained deep learning models (e.g., BERT, GPT-3) learned on large-scale datasets have shown their effectiveness over conventional methods. The big progress is mainly contributed to the representation ability of transformer and its variant architectures. In this paper, we study the low-level computer vision task (e.g., denoising, super-resolution and deraining) and develop a new pre-trained model, namely, image processing transformer (IPT). To maximally excavate the capability of transformer, we present to utilize the well-known ImageNet benchmark for generating a large amount of corrupted image pairs. The IPT model is trained on these images with multi-heads and multi-tails. In addition, the contrastive learning is introduced for well adapting to different image processing tasks. The pre-trained model can therefore efficiently employed on desired task after fine-tuning. With only one pre-trained model, IPT outperforms the current state-of-the-art methods on various low-level benchmarks. Code is available at https://github.com/huawei-noah/Pretrained-IPT and https://gitee.com/mindspore/mindspore/tree/master/model_zoo/research/cv/IPT

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。

↑