semantic segmentation with different dataset
4 次查看(过去 30 天)
显示 更早的评论
how can i use the example in this link
with my own data, means how to creat the pixel label for my data images and when i use the ground turth
0 个评论
回答(1 个)
Vidip
2024-3-26
To create the label pixel for data images which can be used for semantic segmentation using deep learning, you can use ‘Image Labeler’ app as it is an easy way to interactively create a variety of shapes to mark as region of interest (ROI) labels. You can create axis-aligned or rotated rectangular, polyline, pixel, and polygon ROI labels and scene labels in an image or image sequence.
For pixel label, draw pixels to label various classes, such as road or sky, for semantic segmentation. Select the ‘Superpixel’ tool from the toolstrip to create a super pixel grid layout over an image. Select a predefined pixel label definition and then left-click a grid to assign the selected pixel label definition to it. You can remove the labels from ‘superpixel’ grid using the Erase tool to the right of the ‘Superpixel’ tool.
For further information, refer to the documentation links below:
0 个评论
另请参阅
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!