8 commits

Author	SHA1	Message	Date
torzdf	66ed005ef3	Optimize Data Augmentation (#881) ... * Move image utils to lib.image * Add .pylintrc file * Remove some cv2 pylint ignores * TrainingData: Load images from disk in batches * TrainingData: get_landmarks to batch * TrainingData: transform and flip to batches * TrainingData: Optimize color augmentation * TrainingData: Optimize target and random_warp * TrainingData - Convert _get_closest_match for batching * TrainingData: Warp To Landmarks optimized * Save models to threadpoolexecutor * Move stack_images, Rename ImageManipulation. ImageAugmentation Docstrings * Masks: Set dtype and threshold for lib.masks based on input face * Docstrings and Documentation	2019-09-24 12:16:05 +01:00
torzdf	88352b0268	De-Multiprocess Extract (#871) ... * requirements.txt: - Pin opencv to 4.1.1 (fixes cv2-dnn error) * lib.face_detect.DetectedFace: change LandmarksXY to landmarks_xy. Add left, right, top, bottom attributes * lib.model.session: Session manager for loading models into different graphs (for Nvidia + CPU) * plugins.extract._base: New parent class for all extract plugins * plugins.extract.pipeline. Remove MultiProcessing. Dynamically limit batchsize for Nvidia cards. Remove loglevel input * S3FD + FAN plugins. Standardise to Keras version for all backends * Standardize all extract plugins to new threaded codebase * Documentation. Start implementing Numpy style docstrings for Sphinx Documentation * Remove s3fd_amd. Change convert OTF to expect DetectedFace object * faces_detect - clean up and documentation * Remove PoolProcess * Migrate manual tool to new extract workflow * Remove AMD specific extractor code from cli and plugins * Sort tool to new extract workflow * Remove multiprocessing from project * Remove multiprocessing queues from QueueManager * Remove multiprocessing support from logger * Move face_filter to new extraction pipeline * Alignments landmarksXY > landmarks_xy and legacy handling * Intercept get_backend for sphinx doc build # Add Sphinx documentation	2019-09-15 17:07:41 +01:00
torzdf	8e9c901aeb	Bugfix: Extract - Remove BoundingBox object and replace with dict	2019-07-06 09:44:01 +00:00
torzdf	a3294523aa	Centralize cv2 image reading and handle bad filenames	2019-06-02 16:34:43 +01:00
torzdf	26836a4907	Bugfixes ... Facefilter - Variable reference before assignment masks - Facehull initialize to 255.0 instead of 1.0 Alignments tool - Stop tool from ignoring upper case extensions pep8 fixes	2019-05-31 17:48:53 +01:00
torzdf	7b12021dc1	Fixups ... vgg_face: to own module vgg_face: correct output and normalization Extract: more aggressive queue sizes sys_info: Fix cudnn check error on Linux	2019-05-14 16:06:15 +00:00
torzdf	148318778c	Face filter vgg face (#724) ... * Implement extraction pipeline * Face filter to vgg_face. Resume partial model downloads * On-the-fly conversion to extraction pipeline * Move git model ids from get_model to model definition	2019-05-11 18:43:09 +01:00
torzdf	7f53911453	Logging (#541) ... * Convert prints to logger. Further logging improvements. Tidy up * Fix system verbosity. Allow SystemExit * Fix reload extract bug * Child Traceback handling * Safer shutdown procedure * Add shutdown event to queue manager * landmarks_as_xy > property. GUI notes + linting. Aligner bugfix * fix FaceFilter. Enable nFilter when no Filter is supplied * Fix blurry face filter * Continue on IO error. Better error handling * Explicitly print stack trace tocrash log * Windows Multiprocessing bugfix * Add git info and conda version to crash log * Windows/Anaconda mp bugfix * Logging fixes for training	2018-12-04 13:31:49 +00:00