深度学习分割：（1）资料汇总

曾经终败给现在 2022-05-09 07:14 225阅读 0赞

时间关系，备份一些深度学习分割算法相关的博客和介绍文章。

# 分割算法综述 #

介绍了几个比较经典且非常重要的深度学习分割算法。  
原作地址：[https://blog.csdn.net/weixin\_41923961/article/details/80946586][https_blog.csdn.net_weixin_41923961_article_details_80946586]  
更加全面的一篇博客，介绍了FCN, SegNet, DeconvNet, DeepLab, ENet, GCN：[https://blog.csdn.net/YZXnuaa/article/details/79640313][https_blog.csdn.net_YZXnuaa_article_details_79640313]

# FCN #

这是一篇介绍的比较全面且仔细的文章，留以备份，感谢原作认真整理！  
论文地址：[https://people.eecs.berkeley.edu/~jonlong/long\_shelhamer\_fcn.pdf][https_people.eecs.berkeley.edu_jonlong_long_shelhamer_fcn.pdf]  
原作地址：[https://blog.csdn.net/scutjy2015/article/details/74500525][https_blog.csdn.net_scutjy2015_article_details_74500525]  
![在这里插入图片描述][70]

# DeconvNet #

模型的论文名是 DeconvNet: Learning Deconvolution Network for Semantic Segmentation  
论文地址：[https://arxiv.org/abs/1505.04366][https_arxiv.org_abs_1505.04366]  
官方代码：[https://github.com/HyeonwooNoh/DeconvNet\#deconvnet-learning-deconvolution-network-for-semantic-segmentation][https_github.com_HyeonwooNoh_DeconvNet_deconvnet-learning-deconvolution-network-for-semantic-segmentation]  
博客解说：[https://blog.csdn.net/zhangjunhit/article/details/72528610][https_blog.csdn.net_zhangjunhit_article_details_72528610]  
![在这里插入图片描述][70 1]

# SegNet #

SegNet的官网：[http://mi.eng.cam.ac.uk/projects/segnet/\#publication][http_mi.eng.cam.ac.uk_projects_segnet_publication]

![在这里插入图片描述][70 2]

# UNet #

论文官网：[https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/][https_lmb.informatik.uni-freiburg.de_people_ronneber_u-net]  
![在这里插入图片描述][70 3]

## Unet如何改进呢？ ##

改进论文的介绍博客：  
（1）[https://mp.weixin.qq.com/s/ZNNwK1pkL4e0KeYw-UycgA][https_mp.weixin.qq.com_s_ZNNwK1pkL4e0KeYw-UycgA]  
（2）[https://cloud.tencent.com/developer/article/1052565][https_cloud.tencent.com_developer_article_1052565]  
论文地址：[https://arxiv.org/pdf/1801.05746.pdf][https_arxiv.org_pdf_1801.05746.pdf]  
论文代码：[https://github.com/ternaus/TernausNet][https_github.com_ternaus_TernausNet]  
![70 4][]

[https_blog.csdn.net_weixin_41923961_article_details_80946586]: https://blog.csdn.net/weixin_41923961/article/details/80946586
[https_blog.csdn.net_YZXnuaa_article_details_79640313]: https://blog.csdn.net/YZXnuaa/article/details/79640313
[https_people.eecs.berkeley.edu_jonlong_long_shelhamer_fcn.pdf]: https://people.eecs.berkeley.edu/~jonlong/long_shelhamer_fcn.pdf
[https_blog.csdn.net_scutjy2015_article_details_74500525]: https://blog.csdn.net/scutjy2015/article/details/74500525
[70]: /images/20220509/09b645c3c16f4c27b80fe286a1fd4c90.png
[https_arxiv.org_abs_1505.04366]: https://arxiv.org/abs/1505.04366
[https_github.com_HyeonwooNoh_DeconvNet_deconvnet-learning-deconvolution-network-for-semantic-segmentation]: https://github.com/HyeonwooNoh/DeconvNet#deconvnet-learning-deconvolution-network-for-semantic-segmentation
[https_blog.csdn.net_zhangjunhit_article_details_72528610]: https://blog.csdn.net/zhangjunhit/article/details/72528610
[70 1]: /images/20220509/371c4e7a8cbc480f9b9b6b64b4da1f38.png
[http_mi.eng.cam.ac.uk_projects_segnet_publication]: http://mi.eng.cam.ac.uk/projects/segnet/#publication
[70 2]: /images/20220509/754675c229804a02ad3ecc3f05ecb482.png
[https_lmb.informatik.uni-freiburg.de_people_ronneber_u-net]: https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/
[70 3]: /images/20220509/d96dee650a67454aa2a2afde675e5367.png
[https_mp.weixin.qq.com_s_ZNNwK1pkL4e0KeYw-UycgA]: https://mp.weixin.qq.com/s/ZNNwK1pkL4e0KeYw-UycgA
[https_cloud.tencent.com_developer_article_1052565]: https://cloud.tencent.com/developer/article/1052565
[https_arxiv.org_pdf_1801.05746.pdf]: https://arxiv.org/pdf/1801.05746.pdf
[https_github.com_ternaus_TernausNet]: https://github.com/ternaus/TernausNet
[70 4]: /images/20220509/3cddb6964271430da18120cc827b9ff8.png