Deep Interactive Learning
Deep interactive learning or deep interactive segmentation tries to reduce the annotation effort of the expert when annotating new image data. By inferring a refined prediction from an already existing annotation and additional user input, the annotation effort can be reduced.
Resources used
Luo, X., Wang, G., Song, T., Zhang, J., Aertsen, M., Deprest, J., Ourselin, S., Vercauteren, T., & Zhang, S. (n.d.). MIDeepSeg: Minimally Interactive Segmentation of Unseen Objects from Medical Images Using Deep Learning.
Publications
MIDeepSeg: Minimally Interactive Segmentation of Unseen Objects from Medical Images Using Deep Learning
Luo X, Wang G, Song T, Zhang J, Aertsen M, Deprest J, Ourselin S, Vercauteren T, Zhang S - arXiv - 2021
