arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1709.01476cs.CVcs.AIcs.LG
收藏

根据特定的MS COCO类别微调深度CNN模型

Fine-tuning deep CNN models on specific MS COCO categories

German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心);Eötvös Loránd University(罗兰大学)

Daniel Sonntag, Michael Barz, Jan Zacharias, Sven Stauden, Vahid Rahmani, Áron Fóthi, András Lőrincz

摘要:

深度卷积神经网络(CNN)的微调通常是期望的。本文概述了我们公开提供的py-faster-rcnn-ft软件库,可用于在Microsoft Common Objects in Context(MS COCO)数据集的自定义子集上微调VGG_CNN_M_1024模型。例如,我们改进了程序,使用户不必手动在数据集中寻找合适的图像文件,然后可以在演示程序中使用。我们的实现随机选择包含至少一个对象的类别上的模型进行微调的图像。

英文摘要:

Fine-tuning of a deep convolutional neural network (CNN) is often desired. This paper provides an overview of our publicly available py-faster-rcnn-ft software library that can be used to fine-tune the VGG_CNN_M_1024 model on custom subsets of the Microsoft Common Objects in Context (MS COCO) dataset. For example, we improved the procedure so that the user does not have to look for suitable image files in the dataset by hand which can then be used in the demo program. Our implementation randomly selects images that contain at least one object of the categories on which the model is fine-tuned.

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。

↑