RyoKoba
7ccabfa200
translationBot(ui): update translation (Japanese)
...
Currently translated at 68.0% (1304 of 1915 strings)
Co-authored-by: RyoKoba <kobayashi_ryo@cyberagent.co.jp >
Translate-URL: https://hosted.weblate.org/projects/invokeai/web-ui/ja/
Translation: InvokeAI/Web UI
2025-05-30 14:48:41 +10:00
Riccardo Giovanetti
936d59cc52
translationBot(ui): update translation (Italian)
...
Currently translated at 98.9% (1894 of 1915 strings)
Co-authored-by: Riccardo Giovanetti <riccardo.giovanetti@gmail.com >
Translate-URL: https://hosted.weblate.org/projects/invokeai/web-ui/it/
Translation: InvokeAI/Web UI
2025-05-30 14:48:41 +10:00
psychedelicious
fc16fb6099
chore: bump version to v5.13.0rc2
2025-05-30 14:16:33 +10:00
psychedelicious
c848cbc2e3
feat(app): move output annotation checking to run_app
...
Also change import order to ensure CLI args are handled correctly. Had to do this bc importing `InvocationRegistry` before parsing args resulted in the `--root` CLI arg being ignored.
2025-05-30 14:10:13 +10:00
psychedelicious
66fd0f0d8a
feat(ui): warn on unregistered invocation output
2025-05-30 14:10:13 +10:00
psychedelicious
c266f39f06
chore(ui): typegen
2025-05-30 13:36:04 +10:00
psychedelicious
98a44fa4d7
fix(ui): conditional display of message
2025-05-30 13:36:04 +10:00
Mary Hipp
c1d230f961
add support to delete all uncategorized images
2025-05-30 13:36:04 +10:00
Kevin Turner
68108435ae
feat(LoRA): allow LoRA layer patcher to continue past unknown layers
2025-05-30 13:29:02 +10:00
psychedelicious
e121bf1f62
feat(ui): persist sizes of all 4 prompt boxes
2025-05-30 12:36:06 +10:00
psychedelicious
4835c344b3
feat(ui): implement generalized textarea size tracking system
2025-05-30 12:36:06 +10:00
Mary Hipp
a589dec122
store positive prompt textarea height in redux so it persists across refresh
2025-05-30 12:36:06 +10:00
dunkeroni
bc67d5c841
add invert logic to grayscale mask composite
2025-05-30 11:19:37 +10:00
Mary Hipp
f3d5691c04
use onClickGoToModelManager for empty model picker
2025-05-29 11:13:55 -04:00
psychedelicious
b98abc2457
chore(ui): typegen
2025-05-29 13:49:07 +10:00
psychedelicious
7e527ccfb7
feat(api): add validationg for max resize_to on upload endpoint
2025-05-29 13:49:07 +10:00
psychedelicious
e4818b967b
tidy(api): remove benchmark logging
2025-05-29 13:49:07 +10:00
psychedelicious
ce3eede26f
feat(nodes): revised heuristic_resize
...
better handling for smaller image sizes
2025-05-29 13:49:07 +10:00
psychedelicious
d98725c5e9
feat(nodes): use guo-hall thinning
2025-05-29 13:49:07 +10:00
psychedelicious
31a96d2945
feat(ui): use resize on uplaod functionality when creating new canvas from image
2025-05-29 13:49:07 +10:00
psychedelicious
845a321a43
feat(ui): support resize_to when uploading images
2025-05-29 13:49:07 +10:00
psychedelicious
87a44a28ef
chore(ui): typegen
2025-05-29 13:49:07 +10:00
psychedelicious
d5b9c3ee5a
feat(api): support resizing image on upload
2025-05-29 13:49:07 +10:00
psychedelicious
91db136cd1
feat(nodes): much faster heuristic resize utility
...
Add `heuristic_resize_fast`, which does the same thing as `heuristic_resize`, except it's about 20x faster.
This is achieved by using opencv for the binary edge handling isntead of python, and checking only 100k pixels to determine what kind of image we are working with.
Besides being much faster, it results in cleaner lines for resized binary canny edge maps, and has results in fewer misidentified segmentation maps.
Tested against normal images, binary canny edge maps, grayscale HED edge maps, segmentation maps, and normal images.
Tested resizing up and down for each.
Besides the new utility function, I needed to swap the `opencv-python` dep for `opencv-contrib-python`, which includes `cv2.ximgproc.thinning`. This function accounts for a good chunk of the perf improvement.
2025-05-29 13:49:07 +10:00
psychedelicious
c600bc867d
chore: bump version to v5.13.0rc1
2025-05-27 13:30:34 +10:00
Riccardo Giovanetti
f4140dd772
translationBot(ui): update translation (Italian)
...
Currently translated at 98.9% (1890 of 1911 strings)
translationBot(ui): update translation (Italian)
Currently translated at 98.9% (1890 of 1911 strings)
Co-authored-by: Riccardo Giovanetti <riccardo.giovanetti@gmail.com >
Translate-URL: https://hosted.weblate.org/projects/invokeai/web-ui/it/
Translation: InvokeAI/Web UI
2025-05-27 13:18:06 +10:00
psychedelicious
a2d8261d40
feat(ui): canvas scroll scale snap
2025-05-27 13:10:57 +10:00
psychedelicious
bce88a8873
perf(ui): lazy mount scale slider popover
2025-05-27 13:10:57 +10:00
psychedelicious
b37e1a3ad6
feat(ui): do not round scale
...
Makes it a lot smoother, don't think it breaks anything...
2025-05-27 13:10:57 +10:00
psychedelicious
35a088e0a6
perf(ui): optimize <CanvasToolbarScale />
2025-05-27 13:10:57 +10:00
psychedelicious
b936cab039
feat(ui): add computed for stage scale
2025-05-27 13:10:57 +10:00
psychedelicious
34e4093408
fix(ui): revert snapping logic, doesn't work w/ certain input devices
2025-05-27 13:10:57 +10:00
Kent Keirsey
d4c4926caa
Update Compel to 2.1.1 and apply Sentences Split logic
2025-05-26 22:54:15 -04:00
psychedelicious
558c7db055
chore(ui): knipignore InpaintMaskAddButtons
2025-05-27 07:28:47 +10:00
psychedelicious
2ece59b51b
feat(ui): remove unnecessary type casts
2025-05-27 07:28:47 +10:00
psychedelicious
7dbe39957c
feat(ui): bbox rect is always defined, no need for fallback logic
2025-05-27 07:28:47 +10:00
psychedelicious
6fa46d35a5
feat(ui): inpaint mask settings layout
2025-05-27 07:28:47 +10:00
psychedelicious
b2a2b38ea8
feat(ui): split inpaint mask setting selectors to avoid manual memoization
2025-05-27 07:28:47 +10:00
dunkeroni
12934da390
Use Optional instead of Nullable for mask settings
2025-05-27 07:28:47 +10:00
dunkeroni
231bc18188
remove buttons, change denoise limit format
2025-05-27 07:28:47 +10:00
dunkeroni
530cd180c5
chore:ruff
2025-05-27 07:28:47 +10:00
dunkeroni
2a92e7b920
Flux/CogView/SD3 compatible with gradient masks
2025-05-27 07:28:47 +10:00
dunkeroni
019e057e29
chore: typegen
2025-05-27 07:28:47 +10:00
dunkeroni
9aa26f883e
chore: ruff
2025-05-27 07:28:47 +10:00
dunkeroni
3f727e24b1
change default noise level to 0.15
2025-05-27 07:28:47 +10:00
dunkeroni
9e90bf1b20
fix gradient mask broken with flux gen
2025-05-27 07:28:47 +10:00
dunkeroni
db3964797f
clean up comments
2025-05-27 07:28:47 +10:00
dunkeroni
881efbda1b
fix: inpaint breaks when scaled processing
2025-05-27 07:28:47 +10:00
dunkeroni
e9ce2ed5f2
inpaint mask sliders compatible with outpainting
2025-05-27 07:28:47 +10:00
dunkeroni
53ac9eafbf
reuse inpaint image noise seed for caching
2025-05-27 07:28:47 +10:00
