github/DeepFaceLab
1
0
Fork 0
mirror of https://github.com/iperov/DeepFaceLab.git synced 2025-07-07 21:42:08 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions
329 commits 3 branches 2 tags 866 MiB
Commit graph

329 commits

This branch
This branch
All branches
Author SHA1 Message Date
iperov
016791bb6f fix output scale modifier for expanding face 2018-12-13 00:57:55 +04:00
iperov
1bd3a8e0c1 fix error on small mask region size in converter, 
error catching on convert face
 2018-12-12 20:48:30 +04:00
iperov
b52793218e fix 2018-12-08 10:44:46 +04:00
iperov
d11637ec55 limiting max convert subprocessors to 6 2018-12-08 10:12:41 +04:00
iperov
52847bd58f Merge branch 'master' of https://github.com/iperov/DeepFaceLab 2018-12-06 09:32:58 +04:00
iperov
f90b2d231b round created_vram_gb to nearest value. 2018-12-06 09:32:28 +04:00
iperov
a43e93aa0c
added Steve Jobs faceset 2018-12-05 18:20:07 +04:00
iperov
388afc99dc
youtube tutorial for prebuilt DeepFaceLab 2018-12-05 10:29:27 +04:00
iperov
52f2be80d7 print filename if exception occurs while converting 2018-12-04 22:23:02 +04:00
iperov
9040966f90 upgrading to tf 1.11.0 and keras 2.2.4 2018-12-01 22:53:50 +04:00
iperov
5c43f4245e transfercolor via lab converter now implemented by tensorflow-cpu, which is x2 faster than skimage. 
We cannot use GPU for lab converter in converter multiprocesses, because almost all VRAM ate by model process, so even 300Mb free VRAM not enough for tensorflow lab converter.
Removed skimage dependency.
Refactorings.
 2018-12-01 12:11:54 +04:00
iperov
6f0d38d171 converter added --final-image-color-degrade-power - Degrades colors of final image to hide face problems. Valid range [0..100] 2018-11-30 20:08:30 +04:00
iperov
cef6710d24 enhanced masked hist match 2018-11-28 21:07:26 +04:00
iperov
64c3e57f1c added option to converter --output-face-scale-modifier 2018-11-28 20:38:48 +04:00
iperov
2576a411a5
Update main.py 2018-11-21 16:29:48 +04:00
iperov
ef7fdd49a0
Merge pull request #47 from MaksV79/master 
Added options --alpha and --transfercolor
 2018-11-20 09:24:15 +04:00
MaksV79
d47ef23df1
Fix options --alpha and --transfercolor 
Always use params from dialog
 2018-11-19 23:31:06 +03:00
MaksV79
b4af9d3c36
added skimage 2018-11-17 18:20:11 +03:00
MaksV79
6bbb8ad59b
Added option --alpha and --transfercolor 2018-11-17 18:14:17 +03:00
MaksV79
3dcedf6bf3
Added options --alpha and --transfer color 
--alpha will make export png with alpha channel (export only face)
--transfercolor will transfer color from original DST image to fake
 2018-11-17 18:12:38 +03:00
iperov
46eea1c113
Merge pull request #45 from TooMuchFun/keras-2.2.2 
update to Keras 2.2.2
 2018-11-16 18:33:10 +04:00
TooMuchFun
57fd6735a7 update to Keras 2.2.2 2018-11-14 07:22:27 -08:00
iperov
c90d67bec1
Update README.md 2018-11-10 09:59:08 +04:00
iperov
6c4a69e3cd
Update README.md 2018-10-05 13:18:13 +04:00
iperov
4138727466
Merge pull request #20 from andenixa/patch-2 
Landmarks nose drawing fix
 2018-08-24 15:18:01 +04:00
Artem Ivanov
f87ee259b0
Landmarks nose drawing fix 
Fixes nose landmarks drawing if `image_landmarks` are passes as a `numpy.array`
 2018-08-24 14:16:47 +03:00
iperov
f872f07476
Merge pull request #19 from mrsmith0x00/linux_install_fix 
Fix for GNU/Linux install.
 2018-08-24 09:29:29 +04:00
German Vechtomov
99205d6dfd
Update LINUX.md 2018-08-24 05:01:53 +00:00
iperov
07e792e8c5
Merge pull request #16 from NemirovD/master 
Fix for Issue #13
 2018-08-20 09:48:38 +04:00
David
b877367260 Fix for Issue #13 
Swapped + operator for np.concatenate to ensure that the resulting array is 2 by X.
This ensures that we are able to draw the point in the array.
 2018-08-19 17:41:57 -04:00
iperov
61e05f7fdf
Merge pull request #11 from mrsmith0x00/linux 
GNU/Linux installation instructions and helpful bash script
 2018-07-30 16:45:55 +04:00
German Vechtomov
5d3c33898b
Create LINUX.md 2018-07-30 17:38:15 +05:00
German Vechtomov
b7e17be85c
Create main.sh 2018-07-30 17:30:56 +05:00
iperov
a1ff86a6b4 1 2018-07-09 00:09:24 +04:00
iperov
e20527254b
Merge pull request #10 from andenixa/patch-1 
Better face landmark representation
 2018-07-09 00:06:27 +04:00
Artem Ivanov
a8c0613c79
shorter and faster version of draw_landmarks 
- drawing lines via single CV call
 2018-07-07 16:58:50 +03:00
Artem Ivanov
48e281f675
Better face landmark representation 
- fixed nose missing one point
- made AA lines instead of regular ones
- made closed polylines for eyes and mouth
- reduced radius of circles for eyes, nose, and mouth as they obscure lines for small images
 2018-07-07 04:59:17 +03:00
iperov
bda1eed582 upd readme 2018-07-06 13:32:01 +04:00
iperov
a5b9852bfb upd readme 2018-07-03 21:45:20 +04:00
iperov
aab0ab2b08 upd readme 2018-07-03 21:42:48 +04:00
iperov
f11da9acfc
Merge pull request #8 from ChrisTopherTa54321/improve_manual_selection 
Improve manual selection
 2018-07-03 21:40:06 +04:00
iperov
2c74f3bc34 upd readme 2018-07-01 11:54:30 +04:00
iperov
f9fb1351fb update readme 2018-07-01 11:53:55 +04:00
Christopher Throwaway
a74b57d6e5 Improve Manual Extraction process 
Adds improvements to the manual selection process. You can now
click to lock the selection box, and can also go forward and backwards
to modify selections after they have been made. You can also now push
'Q' to skip all remaining images to quickly save the extractions you
have already made.
 2018-06-27 21:32:35 -05:00
Christopher Throwaway
c8c8e8dadc Manual preview now draws 68-pt face landmarks 
It could be difficult to tell if the point cloud was 'correct' or not
when manually fixing a face detection. For 68-point face landmarks, the
facial landmarks are now drawn to make it easier to tell if the face
is correctly detected.
 2018-06-27 21:32:31 -05:00
iperov
46b2aa86da upd readme 2018-06-25 20:38:50 +04:00
iperov
4f9ac71745 upd readme 2018-06-18 18:22:22 +04:00
iperov
08a85a2a5a upd readme 2018-06-15 22:56:28 +04:00
iperov
27893f5981 upd readme 2018-06-09 10:54:27 +04:00
iperov
dec944ad85 upd readme 2018-06-09 10:41:22 +04:00