###
:2018,27(10):226-231
←前一篇   |   后一篇→
本文二维码信息
码上扫一扫!
基于迁移学习的商品图像检测方法
(中国科学技术大学 信息科学技术学院, 合肥 230031)
Product Image Detection Based on Transfer Learning
(School of Information Science and Technology, University of Science and Technology of China, Hefei 230031, China)
摘要
图/表
参考文献
相似文献
本文已被:浏览 342次   下载 276
投稿时间:2018-03-15    修订日期:2018-04-18
中文摘要: 近年来,对象识别方法被应用到多个领域.如人脸检测,车辆检测.然而模型训练所需要的边框标定需要很大的工作量.本文通过基于迁移学习的方法,将物体检测任务迁移到商品检测,且不需要边框标定.本文在分类层和边框回归层之间建立关系层,来学习两种任务之间的关联.本文建立了一个商品数据集,并提出了一种深度学习训练方法,解决了可旋转物体的检测问题.基于Faster RCNN框架,本文提出一种候选选择方法,可以在无边框标定情况下训练商品分类.本文提出的商品检测方法不需要边框标定,而且很容易训练并应用到其它数据集.
Abstract:In recent years, object detection is transferred to other fields, for example, face and vehicle detection. However, the bounding-box labeling is a huge resources cost work. This study solves the problem that transfer object detection task to other domain dataset without bounding-box label. A relationship layer is built to learn the relationship between classification and regression task. In addition, we construct a product dataset, on which rotatable object detection is solved using our training method. A proposal selecting method is proposed for training classification based on faster RCNN framework without bounding-box label. We propose a object detection method without bounding-box annotation. The method is easy to transfer to other datasets and training.
文章编号:     中图分类号:    文献标志码:
基金项目:中科院先导专项课题(XDA06011203)
引用文本:
胡正委,朱明.基于迁移学习的商品图像检测方法.计算机系统应用,2018,27(10):226-231
HU Zheng-Wei,ZHU Ming.Product Image Detection Based on Transfer Learning.COMPUTER SYSTEMS APPLICATIONS,2018,27(10):226-231

用微信扫一扫

用微信扫一扫