语义分割-弱监督学习

CV论文笔记:
https://github.com/ming71/CV_PaperDaily

知乎的文章,比较老,但可以查看先前的工作
https://zhuanlan.zhihu.com/p/23811946
新的
https://www.zhihu.com/question/53263115

该部分总结截止到2019年4月
https://github.com/JackieZhangdx/WeakSupervisedSegmentationList

关于弱监督语义分割代码与文章搜索
https://paperswithcode.com/task/weakly-supervised-semantic-segmentation

ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentation(2016)
https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Lin_ScribbleSup_Scribble-Supervised_Convolutional_CVPR_2016_paper.pdf

Seed, Expand and Constrain Three Principles for Weakly-Supervised Image Segmentation(2016)
https://github.com/kolesman/SEC
https://arxiv.org/pdf/1603.06098.pdf
原则:

  • 种子:给出定位种子
  • 扩展:对定位进行像素级扩展
  • 约束:对边界进行约束

中文博客:
《Seed, Expand and Constrain Three Principles for Weakly-Supervised Image Segmentation》笔记:http://blog.rexking6.top/2018/07/17/%E3%80%8ASeed-Expand-and-Constrain-Three-Principles-for-Weakly-Supervised-Image-Segmentation%E3%80%8B%E7%AC%94%E8%AE%B0/

Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Features(2018)
Weakly-Supervised Semantic Segmentation Network with Deep Seeded Region Growing(2018)
DSRG:code:https://github.com/speedinghzl/DSRG
框架为:caffe

Weakly-and semi-supervised learning of a DCNN for semantic image segmentation(2015)
将像素级标签和图像级标签相结合进行若监督训练,EM-Adapt
paper:https://arxiv.org/pdf/1502.02734.pdf
code:https://github.com/xtudbxk/em-adapt-tensorflow,但该代码表示实现效果只有38%,但论文中效果较好

One-Shot Video Object Segmentation(2017)
对第一视频帧手动分割,后面接续进行自行分割
工程网址:http://people.ee.ethz.ch/~cvlsegmentation/osvos-s/
paper:http://openaccess.thecvf.com/content_cvpr_2017/papers/Caelles_One-Shot_Video_Object_CVPR_2017_paper.pdf

Fast and Accurate Online Video Object Segmentation via Tracking Parts(2018)
paper:https://arxiv.org/abs/1806.02323
code:https://github.com/JingchunCheng/FAVOS
在线训练显存消耗量大,12G
中文解释:https://blog.csdn.net/qq_16761599/article/details/80757611

Learning to Segment Every Thing(2017)(偏监督学习)
paper:https://arxiv.org/pdf/1711.10370.pdf
中文解释:https://www.cnblogs.com/majiale/p/9398590.html

Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relations(2019)
code:https://github.com/jiwoon-ahn/irn,torch
paper:https://arxiv.org/abs/1904.05044
也是生成种子,再繁殖的思想

Weakly- and Semi-Supervised Panoptic Segmentation(2018)
code:https://github.com/qizhuli/Weakly-Supervised-Panoptic-Segmentation, matlab
project: https://qizhuli.github.io/publication/weakly-supervised-panoptic-segmentation/
以目标候选框和目标类别为训练数据集,进行训练

Self-supervised Scale Equivariant Network for Weakly Supervised Semantic Segmentaion(2019)
code:https://github.com/YudeWang/SSENet-pytorch

[论文笔记] 弱监督语义分割 半监督语义分割:https://blog.csdn.net/soulmeetliang/article/details/90261809
Adversarial learning for semi-supervised semantic segmentation
paper
Backtracking ScSPM Image Classifier for Weakly Supervised Top-down Saliency
paper

Weakly Supervised Semantic Segmentation Using Superpixel Pooling Network
paper

  • 0
    点赞
  • 11
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值