'How can i understand the mask picture in Image Segment task?
when i read the tutorial published by tensorflow: https://www.tensorflow.org/tutorials/images/segmentation. I am very confused with the representation of the mask. In the document, it says the Oxford-IIIT Pet dataset's mask picture is marked with {1,2,3}。In processing procedure, input_mask -= 1 is done in order to turn the labels to {0,1,2}.But When I checked the mask numpy(shape is (128,128,1)), it's values consists of other values like 1.1211. It confuses me a lot because i think the numpy should consists of {0,1,2} , it should not include other values.
Please help me with this question! Thanks!
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|