Intellesis Training Options
Selects the set of feature extractors used for training Intellesis segmentation models, see Feature Extractors. Keep in mind that there is no definitive 'correct' selection of parameters. It is advisable to experiment with various parameters for the same image to determine which configuration yields the best results.
|
Parameter |
Description |
|---|---|
|
Basic Features 25 |
A predefined feature set using 25 features, see Basic Features 25. |
|
Basic Features 33 |
A predefined feature set using 33 features, see Basic Features 33. |
|
Deep Features 50 Deep Features 64 Deep Features 70 Deep Features 128 Deep Features 256 |
The complete or reduced feature set from either the 1st, 2nd or 3rd layer of a pre-trained network is used to extract the respective number of features, see Intellesis Deep Features or the respective Zeiss GitHub page. |