• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    迪恩网络公众号

xiaomengyc/SPG: (ECCV2018) Self-produced Guidance for Weakly-supervised Object L ...

原作者: [db:作者] 来自: 网络 收藏 邀请

开源软件名称(OpenSource Name):

xiaomengyc/SPG

开源软件地址(OpenSource Url):

https://github.com/xiaomengyc/SPG

开源编程语言(OpenSource Language):

Python 99.4%

开源软件介绍(OpenSource Introduction):

Self-produced Guidance for Weakly-supervised Object Localization

We train the SPG model on the ILSVRC dataset, and then apply the trained model on video sequences of DAVIS 2016.

Overview of SPG

Train

We finetune the SPG model on the ILSVRC dataset.

cd scripts
sh train_imagenet_full_v5.sh

Test

Download the pretrined model at GoogleDrive(https://drive.google.com/open?id=1EwRuqfGASarGidutnYB8rXLSuzYpEoSM (IMAGENET),https://drive.google.com/open?id=1WfrELBlEoq5WO7gKUv-MLTQ8QHY-2wiX (CUB)).

Use the test script to generate attention maps.

cd scripts
sh val_imagenet_full.sh

Demo

Thanks to Jun Hao for providing the wonderful demos!

Please see the setup_demo.txt for more guidance of setuping up the demos.

Masks are getting better with the proposed easy-to-hard approach.

Citation

If you find this code helpful, please consider to cite this paper:

@inproceedings{zhang2018self,
  title={Self-produced Guidance for Weakly-supervised Object Localization},
  author={Zhang, Xiaolin and Wei, Yunchao and Kang, Guoliang and Yang, Yi and Huang, Thomas},
  booktitle={European Conference on Computer Vision},
  year={2018},
  organization={Springer}
}



鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
热门推荐
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap