Labeling using interactive assisted segmentation

Subject matter regards improving image segmentation or image annotation. A method can include receiving, through a user interface (UI), for each class label of class labels to be identified by the ML model and for a proper subset of pixels of the image data, data indicating respective pixels associated with the class label, partially training the ML model based on the received data, generating, using the partially trained ML model, pseudo-labels for each pixel of the image data for which a class label has not been received, and receiving, through the UI, a further class label that corrects a pseudo-label of the generated pseudo-labels. 12198200_1 (GHMatters) P113259.AU RECEIVE DATA INDICATING RESPECTIVE PIXELS 102 ASSOCIATED WITH EACH CLASS LABEL OF CLASS TRAIN AN ML MODEL BASED ON THE RECEIVED GENERATE, USING THE TRAINED ML MODEL, 106 PSEUDO-LABELS FOR PIXELS FOR WHICH A CLASS LABEL HAS NOT BEEN RECEIVED RECEIVE A FURTHER CL SS LABEL THAT 108 REPLACES A PSEUDO-LABELl OF THE GENERATED PSEUDO-LABELS FURTHER TRAIN THE ML MODEL ON THE CLASS LABELS, FURTHER CLASS LABEL, AND PSEUDO- 110 LABELS FOR WHICH A CLASS LABEL OR FURTHER CLASS LABEL HAS NOT BEEN RECEIVED EXECUTE THE TRAINED ML MODEL ON FURTHER IMAGE DATA TO CLASSIFY EACH PIXEL OF THE 112 FURTHER IMAGE DATA OR TRAIN ANOTHER ML MODEL USING THE IMAGE DATA AND ASSOCIATED CLASSLABELS, FURTHER CLASSLABEL,AND PSEUDOLABELS