/cascaded-faster-rcnn

Primary LanguageJupyter Notebook

  1. read README.md in character-faster-rcnn
  2. read README.md in word-faster-rcnn
  3. read README.md in evaluation 4.a) To use character-faster-rcnn compute the outputs for all files in the dataset (not just the test.txt). The best way to do this is to merge train.txt and test.txt in character-faster-rcnn\DataGeneration\ 4.b) generate images with all overlapping annotations and send them to word-faster-rcnn\DataGeneration\maps\.