* FaceSwap 3 (#1515)
* Update extract pipeline
* Update requirements + setup for nvidia
* Remove allow-growth option
* tf.keras to keras updates
* lib.model.losses - Port + fix all loss functions for Keras3
* lib.model - port initializers, layers. normalization to Keras3
* lib.model.autoclip to Keras 3
* Update mixed precision layer storage
* model file to .keras format
* Restructure nn_blocks to initialize layers in __init__
* Tensorboard
- Trainer: Add Torch compatible Tensorboard callbacks
- GUI event reader remove TF dependency
* Loss logging
- Flush TB logs on save
- Replace TB live iterator for GUI
* Backup models on total loss drop rather than per side
* Update all models to Keras3 Compat
* Remove lib.model.session
* Update clip ViT to Keras 3
* plugins.extract.mask.unet-dfl - Fix for Keras3/Torch backend
* Port AdaBelief to Keras 3
* setup.py:
- Add --dev flag for dev tool install
* Fix Keras 3 syntax
* Fix LR Finder for Keras 3
* Fix mixed precision switching for Keras 3
* Add more optimizers + open up config setting
* train: Remove updating FS1 weights to FS2 models
* Alignments: Remove support for legacy .json files
* tools.model:
- Remove TF Saved Format saving
- Fix Backup/Restore + Nan-Scan
* Fix inference model creation for Keras 3
* Preview tool: Fix for Keras3
* setup.py: Configure keras backend
* train: Migration of FS2 models to FS3
* Training: Default coverage to 100%
* Remove DirectML backend
* Update setup for MacOS
* GUI: Force line reading to UTF-8
* Remove redundant Tensorflow references
* Remove redundant code
* Legacy model loading: Fix TFLamdaOp scalar ops and DepthwiseConv2D
* Add vertical offset option for training
* Github actions: Add more python versions
* Add python version to workflow names
* Github workflow: Exclude Python 3.12 for macOS
* Implement custom training loop
* Fs3 - Add RTX5xxx and ROCm 6.1-6.4 support (#1511)
* setup.py: Add Cuda/ROCm version select options
* bump minimum python version to 3.11
* Switch from setup.cgf to pyproject.toml
* Documentation: Update all docs to use automodapi
* Allow sysinfo to run with missing packages + correctly install tk under Linux
* Bugfix: dot naming convention in clip models
* lib.config: Centralise globally rather than passing as object
- Add torch DataParallel for multi-gpu training
- GUI: Group switches together when generating cli args
- CLI: Remove deprecated multi-character argparse args
- Refactor:
- Centralise tensorboard reading/writing + unit tests
- Create trainer plugin interfaces + add original + distributed
* Update installers
- Add support for different mask centering
- Update legacy alignments to store mask centering
- Bugfix: lib.image ImageIO. Ensure unique queues are created (fixes mask tool when Face is input and an output folder is provided)
* Extract
- Implement aligner re-feeding
- Add extract type to pipeline.ExtractMedia
- Add pose annotation to debug
* Convert
- implement centering
- remove usage of feed and reference face properties
- Remove distributed option from convert
- Force update of alignments file on legacy receive
* Train
- Resize preview image to model output size
- Force legacy centering if centering does not exist in model's state file
- Enable training on legacy face sets
* Alignments Tool
- Update draw to include head/pose
- Remove DFL drop + linting
- Remove remove-frames job
- remove align-eyes option
- Update legacy masks to new extract type
- Exit if attempting to merge version 1.0 alignments files with version 2.0 alignments files
- Re-generate thumbnails on legacy upgrade
* Mask Tool
- Update for new extract + bugfix full frame
* Manual Tool
- Update to new extraction method
- Disable legacy alignments,
- extract box bugfix
- extract faces - size to 512 and center on head
* Preview Tool
- Display based on model centering
* Sort Tool
- Use alignments for sort by face
* lib.aligner
- Add Pose Class
- Add AlignedFace Class
- center _MEAN_FACE on x
- Add meta information with versioning to alignments file
- lib.aligner.get_align_matrix to use landmarks not face
- Refactor aligned faces in lib.faces_detect
* lib.logger
- larger file log padding
* lib.config
- Fix global changeable_items
* lib.face_filter
- Use new extracted face images
* lib.image
- bump thumbnail default size to 96px
* Smart Masks - Training
- Reinstate smart mask training code
- Reinstate mask_type back to model.config
- change 'replicate_input_mask to 'learn_mask'
- Add learn mask option
- Add mask loading from alignments to plugins.train.trainer
- Add mask_blur and mask threshold options
- _base.py - Pass mask options through training_opts dict
- plugins.train.model - check for mask_type not None for learn_mask and penalized_mask_loss
- Limit alignments loading to just those faces that appear in the training folder
- Raise error if not all training images have an alignment, and alignment file is required
- lib.training_data - Mask generation code
- lib.faces_detect - cv2 dimension stripping bugfix
- Remove cv2 linting code
* Update mask helptext in cli.py
* Fix Warp to Landmarks
Remove SHA1 hashing from training data
* Update mask training config
* Capture missing masks at training init
* lib.image.read_image_batch - Return filenames with batch for ordering
* scripts.train - Documentation
* plugins.train.trainer - documentation
* Ensure backward compatibility.
Fix convert for new predicted masks
* Update removed masks to components for legacy models.
- Lint simple_tests.py
- Only reformat alignments file if it exists otherwise change filename
- Update legacy alignments to new format at all stages
- faces_detect.Mask.from_dict - logging format fix
- convert.py fix otf for new pipeline
- cli.py - Add note that masks not used. Revert convert masks
- faces_detect.py - Revert non-extract code
- Add .p and .pickle extensions for serializer
- plugins/extract revert some changes
- scripts/fsmedia - Revert code changes
- Pipeline - cleanup
- Consistant alpha channel stripping (fixes single-process)
- Store landmarks as numpy array
- Code attribution
- Normalize feed face and reference face to 0.0 - 1.0 in convert
- Lock in mask VRAM sized
- Add documentation to plugin_loader
- Update alignments tool to work with new format
- Add new serializers (npy + compressed)
- Remove Serializer option from cli
- Revert get_aligned call in scripts/extract
- Default alignments to compressed
- Size masks to 128px and compress
- Remove mask thresholding/blur from generation code
- Add Mask class to lib/faces_detect
- Revert debug landmarks to aligned face
- Revert non-extraction code to staging version
Bugfix: Fully disable keypress monitor for GUI
Bugfix: Preview - Handle missing alignments file
Config changes:
- Separate plugin defaults into their own files
- Move mask_type to global training config
- Add ability to pass in custom config files
sysinfo,py: CUDNN Check fix for unsupported OSes
cli.py: More useful information
scripts/convert.py: Set processes to 1 if no images are passed in
cv2_dnn aligner: Fix bounding box for negative values
scripts/convert.py: Fix for incorrect frame ranges passed in
* Separate predict and implement pool
* Add check and raise error to multithreading
Box functions to config. Add crop box option.
* All masks to mask module. Refactor convert masks
Predicted mask passed from model. Cli update
* Intesect box with mask and fixes
* Use raw NN output for convert
Use raw mask for face adjustments. Split adjustments to pre and post warp
* Separate out adjustments. Add unmask sharpen
* Set sensible defaults. Pre PR Testing
* Fix queue sizes. Move masked.py to lib
* Fix Skip Frames. Fix GUI Config popup
* Sensible queue limits. Add a less resource intensive single processing mode
* Fix predicted mask. Amend smooth box defaults
* Deterministic ordering for video output
* Video to Video convert implemented
* Fixups
- Remove defaults from folders across all stages
- Move match-hist and aca into color adjustments selectable
- Handle crashes properly for pooled processes
- Fix output directory does not exist error when creating a new output folder
- Force output to frames if input is not a video
* Add Color Transfer adjustment method
Wrap info text in GUI plugin configure popup
* Refactor image adjustments. Easier to create plugins
Start implementing config options for video encoding
* Add video encoding config options
Allow video encoding for frames input (must pass in a reference video)
Move video and image output writers to plugins
* Image writers config options
Move scaling to cli
Move draw_transparent to images config
Add config options for cv2 writer
Add Pillow image writer
* Add gif filetype to Pillow. Fix draw transparent for Pillow
* Add Animated GIF writer
standardize opencv/pillow defaults
* [speedup] Pre-encode supported writers in the convert pool (opencv, pillow)
Move scaling to convert pool
Remove dfaker mask
* Fix default writer
* Bugfixes
* Better custom argparse formatting
torzdf