Skip to content

Latest commit

 

History

History
6 lines (4 loc) · 530 Bytes

README.md

File metadata and controls

6 lines (4 loc) · 530 Bytes

semiauto_annotate

Semi-autocomplete object detection annotations using semi-trained models

Currently only works with the Pascal VOC XML format using LabelImg and Keras RetinaNet.

Once annotations have been predicted, extraneous or innacurate annotations can be adjusted using LabelImg, which can then be used to get higher quality annotations at a much faster speed to continue training of original, semi-trained model.