Skip to content

Instantly share code, notes, and snippets.

@zarzen
Last active August 29, 2015 14:25
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save zarzen/1b08d93d521d9d062484 to your computer and use it in GitHub Desktop.
Save zarzen/1b08d93d521d9d062484 to your computer and use it in GitHub Desktop.
Discussion

resource

cpp implementation

CN24 is a complete semantic segmentation framework using fully convolutional networks. It supports a wide variety of platforms (Linux, Mac OS X and Windows) and libraries (OpenCL, Intel MKL, AMD ACML…) while providing dependency-free reference implementations. The software is developed in the Computer Vision Group at the University of Jena.

这个是用来取super pixel的

这个也是cpp的,还是unsupervise,不过不知道行不行

@Ruochen0715
Copy link

这地就直接这么打字?这样,我看里面还有基于视频的。。咱们需要用公共街景数据集,里面都是一张一张的街景照片,基于视频的分割程序咱们肯定没法用。这样,等你回宿舍了,从这里 http://cbcl.mit.edu/software-datasets/streetscenes/ 把CBCL这个街景数据集下下来,里面都是一张张的街景图片,找到分割代码之后,可以看看哪个比较好用

@zarzen
Copy link
Author

zarzen commented Jul 21, 2015

是的直接打字就可以了。。视频的不能用么?视频不就是连续的图片么、、

@zarzen
Copy link
Author

zarzen commented Jul 21, 2015

https://bitbucket.org/deeplab/deeplab-public/
这个就是基于caffe的,而且比较新

@zarzen
Copy link
Author

zarzen commented Jul 21, 2015

https://sites.google.com/site/uqchang/home/cv-code
这个列表上面有几个可以看一下感觉
不过可能相对比较老

@zarzen
Copy link
Author

zarzen commented Jul 21, 2015

https://www.robots.ox.ac.uk/~tvg/code.php
这个代码列表里有Efficient Salient Region Detection,比较新的代码。但是region detection 是一个东西么?感觉是有点像

@zarzen
Copy link
Author

zarzen commented Jul 21, 2015

http://yann.lecun.com/exdb/publis/pdf/couprie-jmlr-14.pdf
这篇是lecun发的,不过是indoor的ss

@zarzen
Copy link
Author

zarzen commented Jul 21, 2015

http://www.cvlibs.net/links.php?page=Main+page
这也是个列表,似乎还算靠谱

@zarzen
Copy link
Author

zarzen commented Jul 21, 2015

https://github.com/tum-vision/AFS
cpp+cuda+opencv
这个好像不对,是“Automatic Feature Selection is a code framework for the feature selection method introduced in”特征选择用的

@Ruochen0715
Copy link

视频的肯定没法用,视频是连续的图片没错,但是街景数据集里的图片并不连续啊。。。。。

@zarzen
Copy link
Author

zarzen commented Jul 29, 2015

"数据集有这样几个都比较好用,KITTI,CBCL StreetScenes Challenge Framework,谷歌街景的一个数据集"

@zarzen
Copy link
Author

zarzen commented Aug 9, 2015

deeplab的caffe model没法直接用原生的caffe编译出来的运行。
应该是因为它们加上了自己的代码字段。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment