semantic segmentation with deep learning

Question

Salma Hassan 2018-12-27

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/437336-semantic-segmentation-with-deep-learning

回答： Johanna Pingel 2019-2-6

i have a confused between pixel label and ground truth , are there similar when using for training SegNet code . if not how can i create the pixellabels , and are 50 images enough to create pixellabels

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Johanna Pingel 2019-2-6

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/437336-semantic-segmentation-with-deep-learning#answer_359783

Check out a few links to learn more about pixel labeling and semantic segmentation:https://www.mathworks.com/videos/semantic-segmentation-overview-1510858047780.html?s_tid=srchtitle

You're going to want the image labeler to do the labeling: https://www.mathworks.com/help/vision/ref/imagelabeler-app.html

The number of images really depends on the complexity of the problem. I was able to do a quick semantic segmentation with only 100 images because the images were so simple: https://blogs.mathworks.com/deep-learning/2018/11/15/sudoku-solver-image-processing-and-deep-learning/