Skip to content
Snippets Groups Projects

Updated changelogs of Bob pad.face, ip.qualitymeasure, bio.video and bio.vein

Merged Updated changelogs of Bob pad.face, ip.qualitymeasure, bio.video and bio.vein
updatelog_oleg into updatelog
Merged Olegs NIKISINS requested to merge updatelog_oleg into updatelog
Compare
and
1 file
+ 66
− 239
Compare changes
  • Side-by-side
  • Inline
release/changelog_since_last_release.txt
+ 66
− 239
@@ -171,14 +171,10 @@
- Make the tests less relaxed
* bob.ip.qualitymeasure
* patch
- added ci config file
- added conda yaml config
- updated README
- removed ref to bootstrapbuildout
- updated doc conf
- removed docopt requirement
- small fixes
- Added an optional reference image to Galbally IQM
* Migrated to conda based CI
* Updates in the README
* Removed bootstrap-buildout functionality
* Added an optional reference image to Galbally IQM extractor
* bob.ip.skincolorfilter
* minor
* Migrate to conda based CI
@@ -447,143 +443,49 @@
* bob.bio.video!20 Docs url fix
* bob.bio.video!19 Python36 & docs fixes
* bob.bio.video!18 Removed`Algorithm.read_probe`
* patch
- Add as_array to frame containers
- Fixed tuple indexing bug in youtube db load function
- Added unit test for the load method of youtube database
- Updated manifest to use jpg files for tests
- Fixed the unit test connecting it with low level db interface]
- Fixed the unit test of youtube load method
- Make the original_directory and annotation_directory a property
- move the _db initialization before calling super
- pass kwargs in the base file class
* Migrate to conda based CI
- [conda] Added test requirements
- Fixed the index.rst to match joint docs requirements
- Added quality_assessment_function to preprocessor Wrapper
- Fixed the typo
- Addressing missing frame annotations
- Fixed dict.has_key issue for Python 3
- Reverted quality assessment function
- Changed missing annotations conditional check + unit tests
- Moved condition to one line, removed empty dictionary check
- Fixes bob.bio.video#11
- Returning none insted of empty FrCont in preproc Wrapper
- Make missing annotations None instead of empty dict
- Add a failsafevideo wrapper
- use ordered dict
- Add a video annotator
- Finalize video annotators and add docs and tests
- Fix import order
- use six to check for string types
- Account for empty frame containers too
- Account for failed extractor
- remove unused import thanks to my linter and the way I imported it
- mention min_face_size_validator for a possible input
- Validators should handle empty annotations too
- Fix sphinx warning
* minor
* Add function as_array() to FrameContainer
* Fixed youtube db load function
* The original_directory and annotation_directory are now a property
* Migrated to conda based CI
* Package now added to joint documentation
* Missing annotations are now handled in the Preprocessor Wrapper
* Updated unit tests addressing missing annotation cases
* Preprocessor Wrapper now able to return None, allowing --missing-files functionality
* Added failsafevideo wrapper
* Added Video annotator
* bob.bio.vein
* v1.0.0 (Oct 20, 2017 13:29)
* Initial release
- Add bob.db.fv3d requirement
- Minor fix to logging setup
- Implement no-cropping option
- Improve sample viewing script
- Add blame script
- Implement empty annotations() method to satisfy bob.bio.base
- Add support for 3D Fingervein Database
- Re-estructured Preprocessor to simplify experimentation and configuration
- Add matlab reference generation and readme
- Remove trailing spaces
- Improve comments
- Add test for 3D fingervein database
- More preprocessor changes
- Use scipy directly where possible
- Fixes to Maximum Curvature implementation; Improved references & documentation
- Update manifest
- Speeds-up the evaluation of vein probabilities
- Missing import
- Avoid warnings
- Ensures input image has dtype==float64 before anything else
- Fix loading of fv3d fingers
- Use 'crop' instead of 'cropper'
- Improve inheritance; Fix FixedMask implementation
- If sample does not have roi, then returns only image
- Add grid resources and explanation on how to run
- Implement optional pre-cropping stage on preprocessor
- Test cropping with annotated images
- Use cropped images on next stages
- More tests for preprocessor specific stages
- Reflected changes from copper->crop renaming
- Don't need to do poly_to_mask() anymore
- Improve comments on TomesLeeMask
- Use padder by default on stock configurations
- Fix missing transposition on port
- Implemented normalized-cross-correlation algorithm
- Remove ipdb break point
- Fix doc error
- Add mlp trainer for watershed masker
- Add mlp trainer for watershed masker
- Add watershed mask detector
- Remove useless mask(); Update comments
- Fix score splitting
- Fix indentation
- Allow to view sample using a database name
- Fixes to marker detector
- Add ON's miura-match with rotation support
- Load only what is annotated (allow flexibility)
- Support view mode w/o extracted veins
- Add watershed preprocessing debug-aid script
- Printout metrics as they are evaluated
- Allow separate thresholds for foreground and background markers; Allow external…
- Use simpler validation
- Add script to help debugging watershedding
- Fix spelling
- Add watershedding link
- Simplify baseline, Add watershed NN training
- Fix link
- Update comment
- Set better defaults and use them on stock configuration
- Added the HLDI for the putvein database
- Added the HLDI for the putvein database
- Update documentation to include 3D-fingervein pointers
- Fix installation directives
- Remove database interfaces from runtime
- Added the HLDI for the putvein database
- Added the doc and tests and config for the putvein database
- Fixed the HLDI of the putvein db - conversion of group names
- Added bob.ip.color to requirements.txt
- Fix loading of binarised images
- Option to save figures discretely
- Include databases on bulidout configurations
- Fix Hamming Distance module
- Make HammingDistance class plus add test
- Improve documentation in HammingDistance class
- Update baseline results for rlt, mc, and wld
- Remove Oleg's implementation of rotation-invariant matching (before release,…
- Update README links [ci skip]
- Complete design and doc of HammingDistance
- Add missing dev packages
- Reset preliminary version to 1.0.0a0
- update the table for verafinger/nom.
- Minor typo fix [ci skip]
* Added bob.db.fv3d
* Updated sample viewing script
* Added blame script
* Re-estructured Preprocessor to simplify experimentation and configuration
* Unit tests for 3D fingervein database
* Fixes to Maximum Curvature implementation; Improved references & documentation
* Speeds-up the evaluation of vein probabilities
* Updated functionality: If sample does not have ROI, then returns only image
* Updated documentation: on grid resources and explanation on how to run
* Added optional pre-cropping stage on preprocessor
* Implemented normalized-cross-correlation algorithm
* Added MLP trainer for watershed masker
* Added watershed mask detector
* Functionality allowing to view sample using a database name
* Added Miura-match with rotation support
* Support for view mode w/o extracted veins
* Added watershed preprocessing debug-aid script
* Added High Level Database Interface for putvein database
* Added the doc and tests and config for the putvein database
* HammingDistance class plus test
* Update documentation on baseline results for rlt, mc, and wld
* Remove Oleg's implementation of rotation-invariant matching
* v1.0.1 (Oct 20, 2017 16:04)
* Update `MANIFEST.in` to include necessary package files
- Update MANIFEST to include essential files
* patch
- updated the whole table. The numbers are very different for the new algorithms.
- Port to new CI (closes bob.bio.vein#16)
- Try a fix for the scikit image dep
- Try another fix for the scikit image dep
- Update utfvp interface
- Remove uncited ref
- Remove uncited ref
- Removes more uncited references
- Enable compilation of docs w/o latex
- Adds bob.bio.vein sphinx ref
- Fixed the index.rst to match joint docs requirements
- Adaptations to allow vulnerability analysis
- Catch-up with changes to verafinger
* Ported to new CI
* Updated utfvp interface
* Fixed the index.rst to match joint docs requirements
* Adaptations to allow vulnerability analysis
* bob.db.voxforge
* v2.2.0 (Sep 24, 2017 11:39)
* patch
@@ -641,106 +543,31 @@
- Fixed convert_frame_cont_to_array addressing missing annotations
- Add a predictions algorithm useful for deeplearning
* bob.pad.face
* patch
- Added MIFS database
- bugfix: array indices fixed
- Added face bounding boxes to tree and cleaned database path
- Updated documentation
- documentation fix
- documentation fix
- Fixed documentation
- bugfix: array indices fixed
- documentation fix
- documentation fix
- Updated documentation
- Added face bounding boxes to tree and cleaned database path
- Fixed documentation
- Cleaning
- Merge
- Fixed merge errors
- Updated documentation
- Updated documentation
- Updated the MANIFEST.in to make annotations available for MIFS db
- Added entry point for VideoSparseCoding preprocessor amd initial version of…
- Addded conversion of sparse codes to Frame Container
- Added sparse encoding part to VideoSparseCoding preprocessor
- Added a part of VideoSparseCoding preprocessor
- Moved histogram extraction to preprocessor VideoSparseCoding
- Added entry points for VideoHistOfSparseCodes extractor
- Fixed an indentation issue in the doc of VideoSparseCoding
- Updated the main doc on anomaly detection PAD for Aggregated Db, some doc for…
- Optimized the VideoSparseCoding for memory useage
- Added the reconstruction error option to VideoSparseCoding preprocessor
- Removed temp stuff from setup.py, cleaning of sporse coding algorithms
- Fixed the path bug in the video_sparse_coding config file
- Added a local face detection option to the VideoFaceCrop class
- Updated requirements.txt with dlib facedetector
- Added unit tests for classes used in the IQM-GMM pad algorithm
- Fixed the None case in face detection utils
- Fixed no face detected issue in VideoFaceCrop
- Modified the grid config requesting less memory
- Added optional face detectors, also empty files will not be saved in the preprocessor
- Fixes unit tests after update of bob.ip.dlib
- Fixed the unit test to match the latets updates in bob.ip.dlib
- Removed a dependency from bob.ip.mtcnn in face detection utils
- Added facelandmark detection in utilsbob.pad.face!
- unit tests
- test
- Added mtcnn along with dlib.
- Added mtcnn/dlib detectors
- Changed function name + some minor modifications
- adapt svm algorithm to other modalities, like voice
- adapt one class gmm to other modalities
- Added a grandtest-train-eval protocol to the aggregated db
- Added grandtest-mobio protocol to the aggregated db
- Minor typo fix in the doc of Aggregated db
- pep8 formatting done by yapf -ri .
- Accept frame_selector in load methods
- remove bootstrap-buildout.py
- Add a base pad file for all face pad databases
- Use all frames. Improve docs
- The inital version of grandtest-train-eval-<num_train_samples> protocol for Aggregated db
- Fixed the None issue in grandtest-train-eval-<num_train_samples> protocol
- Add frame processing functionality
- MethodType in Python 3 takes 2 arguments only
- Make the original_directory and annotation_directory a property
- db not _db
- Implement frame by frame loading in replay and replay-mobile
- use bob.extension.rc instead of None
* Migrate to conda based CI
- Moved algos to bob.pad.base and renamed them
- updating conda recipe
- removing console scripts entry point, since there are no scripts
- removing console scripts entry point, since there are no scripts
- updating conda recipe
- updating conda recipe
- fixing test_video_face_crop and conda recipe
- Changes in conda recipe according to Anirs comments
- Added mobio db to conda recipe, fixed unit tests
- Print the assertion value
- Fixed the nose tests for VideoFaceCrop
- Fixed the pointer to Algos documentation, now in bob.pad.base
- Changed host/run requirements
- Fixed the index.rst to match joint docs requirements
- Removed duplicating db configs, issue bob.pad.face#18
- Removed VideoQualityMeasure, replaced with Wraper
- removed VideoLBPHistogram, replaced with Wrapper
- Removed unused class VideoDataLoader, closes bob.pad.face#20
- Added alignment option to normalize_image_size_in_grayscale in ImageFaceCrop
- Changed hard-coded constants into face size relative in FaceEyesNorm call
- Moved some class methods to functions, WIP
- Minor modifications in ImageFaceCrop, WIP
- Added initial version of FaceCropAlign, which is refactor of ImageFaceCrop
- Fixed the bug in min_face_size check
- Removed VideFaceCrop and ImageFaceCrop, replaced with FaceCropAlign+Wrapper
- Removed face_detection_utils, updated unit tests
- Added underscore to preproc configs to resolve warning when running baselines
- Removed deleted utils from docs
- Added a FrameSelector to the load() in Aggregated Db HLDI
- Changed the default FrameSelector in preproc configs to select all frames
- Add a low-intensive grid option named small
- Use our common right eye and left eye names instead
- Fix the tests too
* major
* Added High Level Database Interface (HLDI) for MIFS database
* Annotations are added for MIFS database
* Added a VideoSparseCoding preprocessor
* Updated the documentation on anomaly detection-based PAD using Aggregated Db
* Local face detection option was added to VideoFaceCrop class, using dlib and mtcnn
* Unit tests for the classes used in IQM-GMM pad algorithm
* Added landmark detection utils
* Algorithms/classifiers are adapted to other non-face modalities
* New protocol for Aggregated Database: grandtest-train-eval
* New protocol for Aggregated Database: grandtest-mobio
* Frame Selector is now acceptable in HLDIs of all databases
* Removed bootstrap-buildout.py functionality
* Added A base PAD file class
* New protocol for Aggregated Database: grandtest-train-eval-<num_train_samples>
* bob.extension.rc used in HLDIs
* Migrated to conda-based CI
* ALL algorithms are moved to bob.pad.base package
* Fixed algorithms documentation links, pointing to bob.pad.base now.
* The documentation of the package now linked to joint documentation properly.
* Refactoring: duplicated database configs are removed
* Refactoring: removed VideoQualityMeasure, replaced with Wrapper class
* Refactoring: removed VideoLBPHistogram, replaced with Wrapper class
* Refactoring: removed unused VideoDataLoader
* Refactoring: created FaceCropAlign, replaces ImageFaceCrop + VideFaceCrop using Wrapper
* bob.db.oulunpu
* patch
- Initial commit