### 1. **DeepFaceLab**
DeepFaceLab은 현재 가장 널리 사용되고 있는 딥페이크 소프트웨어 중 하나로, 특히 영화 및 비디오 콘텐츠 제작에서 많이 활용됩니다. 얼굴 스왑, 얼굴 정합, 고해상도 합성과 같은 매우 정밀한 작업을 수행할 수 있으며, 다양한 딥러닝 모델을 제공하여 사용자 맞춤형 설정이 가능합니다.
### 2. **Faceswap**
Faceswap은 사용자가 얼굴을 매우 정밀하게 교환할 수 있도록 해주는 오픈 소스 프로젝트입니다. 이 프로젝트는 다양한 전문가들과 영화 및 미디어 제작자들 사이에서 인기를 끌고 있으며, 꾸준한 업데이트와 안정적인 지원을 통해 고품질의 결과물을 제공합니다.
### 3. **Avatarify**
이 프로젝트는 실시간 얼굴 표정과 동작을 다른 얼굴에 전이시키는 기능을 제공하며, 스트리밍 및 영화 촬영에서 실시간으로 얼굴 합성을 하는 데 유용합니다. 특히 모션 캡처와 결합하여 매우 자연스러운 합성이 가능합니다.
### 4. **First Order Motion Model**
First Order Motion Model은 주로 애니메이션과 짧은 비디오 클립을 만드는 데 사용되며, 인물의 동작을 매우 정밀하게 전이시킬 수 있습니다. 영화 제작에서는 특정 장면에서 배우의 표정을 다른 얼굴로 변경하거나 애니메이션 캐릭터에 생동감을 불어넣는 데 유용합니다.
### 5. **Thin-Plate Spline Motion Model**
이 프로젝트는 얼굴과 상체의 복잡한 움직임을 추적하여 다른 이미지 또는 비디오에 적용할 수 있습니다. 고해상도의 얼굴 합성 및 정밀한 모션 전이를 필요로 하는 영화 제작에 적합합니다.
### 상업적 솔루션
이외에도 **NVIDIA's GANimation**과 같은 상업적 솔루션들은 영화 제작에 고성능의 딥페이크 기술을 제공하며, 상용 라이센스를 통해 좀 더 안정적이고 고품질의 결과물을 얻을 수 있습니다.
### 주의사항
딥페이크 기술은 높은 계산력과 전문 지식이 필요하며, 적절한 하드웨어 환경이 마련되어야 최고의 성능을 발휘할 수 있습니다. 또한 이러한 기술의 남용 방지와 윤리적 문제 해결을 위해 법적 규제와 정확한 사용자 가이드라인을 따르는 것이 중요합니다.
이러한 프로젝트들은 각각의 특성과 장점이 다르기 때문에, 영화 제작의 특정 요구사항과 목적에 따라 적절한 도구를 선택하는 것이 중요합니다.
import tensorflow as tf
print(tf.__version__)
2.1.0
generator_dataset_op.cc:103] Error occurred when finalizing GeneratorDataset iterator:
https://github.com/tensorflow/tensorflow/issues/37515
Tensorflow 2.1 Error “when finalizing GeneratorDataset iterator” - a memory leak? · Issue #37515 · tensorflow/tensorflow
Reopening of issue #35100, as more and more people report to still have the same problem: Problem description I am using TensorFlow 2.1.0 for image classification under Centos Linux. As my image tr...
github.com
(tf-gpu) C:\Users\joe>pip install tensorflow==2.2.0rc3
Collecting tensorflow==2.2.0rc3
Downloading https://files.pythonhosted.org/packages/af/b6/c634218cd4602e906a922fe8b372d582e29624358f2e997aa1cd097164bf/tensorflow-2.2.0rc3-cp37-cp37m-win_amd64.whl (459.2MB)
|████████████████████████████████| 459.2MB 42kB/s
Requirement already satisfied: grpcio>=1.8.6 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (1.16.1)
Requirement already satisfied: numpy<2.0,>=1.16.0 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (1.17.3)
Requirement already satisfied: wheel>=0.26; python_version >= "3" in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (0.33.6)
Requirement already satisfied: six>=1.12.0 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (1.13.0)
Collecting h5py<2.11.0,>=2.10.0
Requirement already satisfied: opt-einsum>=2.3.2 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (3.1.0)
Requirement already satisfied: wrapt>=1.11.1 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (1.11.2)
Requirement already satisfied: absl-py>=0.7.0 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (0.8.1)
Requirement already satisfied: keras-preprocessing>=1.1.0 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (1.1.0)
Collecting tensorboard<2.3.0,>=2.2.0
Collecting tensorflow-estimator<2.3.0,>=2.2.0rc0
Requirement already satisfied: google-pasta>=0.1.8 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (0.1.8)
Requirement already satisfied: protobuf>=3.8.0 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (3.10.1)
Collecting astunparse==1.6.3
Collecting gast==0.3.3
Collecting scipy==1.4.1; python_version >= "3"
Requirement already satisfied: termcolor>=1.1.0 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorflow==2.2.0rc3) (1.1.0)
Collecting google-auth-oauthlib<0.5,>=0.4.1
Requirement already satisfied: markdown>=2.6.8 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow==2.2.0rc3) (3.1.1)
Collecting tensorboard-plugin-wit>=1.6.0
Downloading https://files.pythonhosted.org/packages/b6/85/5c5ac0a8c5efdfab916e9c6bc18963f6a6996a8a1e19ec4ad8c9ac9c623c/tensorboard_plugin_wit-1.7.0-py3-none-any.whl (779kB)
|████████████████████████████████| 788kB 6.4MB/s
Requirement already satisfied: werkzeug>=0.11.15 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow==2.2.0rc3) (0.16.0)
Collecting requests<3,>=2.21.0
Collecting google-auth<2,>=1.6.3
Requirement already satisfied: setuptools>=41.0.0 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from tensorboard<2.3.0,>=2.2.0->tensorflow==2.2.0rc3) (42.0.1.post20191125)
Collecting requests-oauthlib>=0.7.0
Requirement already satisfied: certifi>=2017.4.17 in j:\anaconda3\envs\tf-gpu\lib\site-packages (from requests<3,>=2.21.0->tensorboard<2.3.0,>=2.2.0->tensorflow==2.2.0rc3) (2020.6.20)
Collecting chardet<4,>=3.0.2
Collecting urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1
Collecting idna<3,>=2.5
Downloading https://files.pythonhosted.org/packages/a2/38/928ddce2273eaa564f6f50de919327bf3a00f091b5baba8dfa9460f3a8a8/idna-2.10-py2.py3-none-any.whl (58kB)
|████████████████████████████████| 61kB ...
Collecting rsa<5,>=3.1.4; python_version >= "3"
Collecting pyasn1-modules>=0.2.1
Collecting cachetools<5.0,>=2.0.0
Collecting oauthlib>=3.0.0
Collecting pyasn1>=0.1.3
ERROR: tensorboard 2.2.2 has requirement grpcio>=1.24.3, but you'll have grpcio 1.16.1 which is incompatible.
Installing collected packages: h5py, pyasn1, rsa, pyasn1-modules, cachetools, google-auth, chardet, urllib3, idna, requests, oauthlib, requests-oauthlib, google-auth-oauthlib, tensorboard-plugin-wit, tensorboard, tensorflow-estimator, astunparse, gast, scipy, tensorflow
Found existing installation: h5py 2.9.0
Uninstalling h5py-2.9.0:
Successfully uninstalled h5py-2.9.0
Found existing installation: tensorboard 2.0.0
Uninstalling tensorboard-2.0.0:
Successfully uninstalled tensorboard-2.0.0
Found existing installation: tensorflow-estimator 2.0.0
Uninstalling tensorflow-estimator-2.0.0:
Successfully uninstalled tensorflow-estimator-2.0.0
Found existing installation: gast 0.2.2
Uninstalling gast-0.2.2:
Successfully uninstalled gast-0.2.2
Found existing installation: scipy 1.3.1
Uninstalling scipy-1.3.1:
Successfully uninstalled scipy-1.3.1
Found existing installation: tensorflow 2.0.0
Uninstalling tensorflow-2.0.0:
Successfully uninstalled tensorflow-2.0.0
Successfully installed astunparse-1.6.3 cachetools-4.1.1 chardet-3.0.4 gast-0.3.3 google-auth-1.18.0 google-auth-oauthlib-0.4.1 h5py-2.10.0 idna-2.10 oauthlib-3.1.0 pyasn1-0.4.8 pyasn1-modules-0.2.8 requests-2.24.0 requests-oauthlib-1.3.0 rsa-4.6 scipy-1.4.1 tensorboard-2.2.2 tensorboard-plugin-wit-1.7.0 tensorflow-2.2.0rc3 tensorflow-estimator-2.2.0 urllib3-1.25.9
(tf-gpu) C:\Users\joe>
muellerdo commented on 9 May •
edited
| Update: Tensorflow 2.2.0 does not support Keras Data Generators for validation, anymore. I'm probably have to rework the complete Data Generator of MIScnn into TF datasets... Any suggestions are welcome. Source: https://www.tensorflow.org/api_docs/python/tf/keras/Model#fit |
(tf-gpu) C:\Users\joe>conda update --all
Collecting package metadata (current_repodata.json): done
Solving environment: done
## Package Plan ##
environment location: J:\Anaconda3\envs\tf-gpu
The following packages will be downloaded:
package | build
---------------------------|-----------------
cudnn-7.6.5 | cuda10.0_0 164.2 MB
pyopenssl-19.1.0 | py37_0 91 KB
pysocks-1.7.1 | py37_0 31 KB
python-3.7.7 | h81c818b_4 14.3 MB
qtpy-1.9.0 | py_0 38 KB
------------------------------------------------------------
Total: 178.6 MB
The following NEW packages will be INSTALLED:
blinker pkgs/main/win-64::blinker-1.4-py37_0
brotlipy pkgs/main/win-64::brotlipy-0.7.0-py37he774522_1000
cachetools pkgs/main/noarch::cachetools-4.1.0-py_1
cffi pkgs/main/win-64::cffi-1.14.0-py37h7a1dbc1_0
chardet pkgs/main/win-64::chardet-3.0.4-py37_1003
click pkgs/main/noarch::click-7.1.2-py_0
cryptography pkgs/main/win-64::cryptography-2.9.2-py37h7a1dbc1_0
google-auth pkgs/main/noarch::google-auth-1.14.1-py_0
google-auth-oauth~ pkgs/main/noarch::google-auth-oauthlib-0.4.1-py_2
idna pkgs/main/noarch::idna-2.10-py_0
importlib-metadata pkgs/main/win-64::importlib-metadata-1.7.0-py37_0
oauthlib pkgs/main/noarch::oauthlib-3.1.0-py_0
packaging pkgs/main/noarch::packaging-20.4-py_0
prompt-toolkit pkgs/main/noarch::prompt-toolkit-3.0.5-py_0
pyasn1 pkgs/main/noarch::pyasn1-0.4.8-py_0
pyasn1-modules pkgs/main/noarch::pyasn1-modules-0.2.7-py_0
pycparser pkgs/main/noarch::pycparser-2.20-py_0
pyjwt pkgs/main/win-64::pyjwt-1.7.1-py37_0
pyopenssl pkgs/main/win-64::pyopenssl-19.1.0-py37_0
pyparsing pkgs/main/noarch::pyparsing-2.4.7-py_0
pysocks pkgs/main/win-64::pysocks-1.7.1-py37_0
qtpy pkgs/main/noarch::qtpy-1.9.0-py_0
requests pkgs/main/noarch::requests-2.24.0-py_0
requests-oauthlib pkgs/main/noarch::requests-oauthlib-1.3.0-py_0
rsa pkgs/main/noarch::rsa-4.0-py_0
tensorboard-plugi~ pkgs/main/noarch::tensorboard-plugin-wit-1.6.0-py_0
urllib3 pkgs/main/noarch::urllib3-1.25.9-py_0
win_inet_pton pkgs/main/win-64::win_inet_pton-1.1.0-py37_0
The following packages will be REMOVED:
more-itertools-7.2.0-py37_0
The following packages will be UPDATED:
absl-py 0.8.1-py37_0 --> 0.9.0-py37_0
backcall pkgs/main/win-64::backcall-0.1.0-py37~ --> pkgs/main/noarch::backcall-0.2.0-py_0
bleach pkgs/main/win-64::bleach-3.1.0-py37_0 --> pkgs/main/noarch::bleach-3.1.5-py_0
colorama pkgs/main/win-64::colorama-0.4.1-py37~ --> pkgs/main/noarch::colorama-0.4.3-py_0
cudnn 7.6.4-cuda10.0_0 --> 7.6.5-cuda10.0_0
decorator 4.4.1-py_0 --> 4.4.2-py_0
google-pasta 0.1.8-py_0 --> 0.2.0-py_0
grpcio 1.16.1-py37h351948d_1 --> 1.27.2-py37h351948d_0
h5py 2.9.0-py37h5e291fa_0 --> 2.10.0-py37h5e291fa_0
icu 58.2-ha66f8fd_1 --> 58.2-ha925a31_3
importlib_metadata pkgs/main/win-64::importlib_metadata-~ --> pkgs/main/noarch::importlib_metadata-1.7.0-0
intel-openmp 2019.4-245 --> 2020.1-216
ipykernel 5.1.3-py37h39e3cac_0 --> 5.3.0-py37h5ca1d4c_0
ipython 7.9.0-py37h39e3cac_0 --> 7.16.1-py37h5ca1d4c_0
jedi 0.15.1-py37_0 --> 0.17.1-py37_0
jinja2 2.10.3-py_0 --> 2.11.2-py_0
jupyter_client pkgs/main/win-64::jupyter_client-5.3.~ --> pkgs/main/noarch::jupyter_client-6.1.3-py_0
jupyter_console pkgs/main/win-64::jupyter_console-6.0~ --> pkgs/main/noarch::jupyter_console-6.1.0-py_0
jupyter_core 4.6.1-py37_0 --> 4.6.3-py37_0
libprotobuf 3.10.1-h7bd577a_0 --> 3.12.3-h7bd577a_0
libsodium 1.0.16-h9d3ae62_0 --> 1.0.18-h62dcd97_0
mkl 2019.4-245 --> 2020.1-216
mkl_fft 1.0.15-py37h14836fe_0 --> 1.1.0-py37h45dec08_0
mkl_random 1.1.0-py37h675688f_0 --> 1.1.1-py37h47e9c7a_0
nbformat pkgs/main/win-64::nbformat-4.4.0-py37~ --> pkgs/main/noarch::nbformat-5.0.7-py_0
notebook 6.0.2-py37_0 --> 6.0.3-py37_0
numpy 1.17.3-py37h4ceb530_0 --> 1.18.5-py37h6530119_0
numpy-base 1.17.3-py37hc3f5095_0 --> 1.18.5-py37hc3f5095_0
pandoc 2.2.3.2-0 --> 2.9.2.1-0
parso 0.5.1-py_0 --> 0.7.0-py_0
pip 19.3.1-py37_0 --> 20.1.1-py37_1
prometheus_client 0.7.1-py_0 --> 0.8.0-py_0
prompt_toolkit 2.0.10-py_0 --> 3.0.5-0
protobuf 3.10.1-py37h33f27b4_0 --> 3.12.3-py37h33f27b4_0
pygments 2.4.2-py_0 --> 2.6.1-py_0
pyrsistent 0.15.6-py37he774522_0 --> 0.16.0-py37he774522_0
python 3.7.5-h8c8aaf0_0 --> 3.7.7-h81c818b_4
pywin32 223-py37hfa6e2cd_1 --> 227-py37he774522_1
pywinpty 0.5.5-py37_1000 --> 0.5.7-py37_0
pyzmq 18.1.0-py37ha925a31_0 --> 19.0.1-py37ha925a31_1
qtconsole 4.6.0-py_0 --> 4.7.5-py_0
scipy 1.3.1-py37h29ff71c_0 --> 1.5.0-py37h9439919_0
setuptools 42.0.1-py37_0 --> 47.3.1-py37_0
six pkgs/main/win-64::six-1.13.0-py37_0 --> pkgs/main/noarch::six-1.15.0-py_0
sqlite 3.30.1-he774522_0 --> 3.32.3-h2a8f88b_0
tensorboard 2.0.0-pyhb38c66f_1 --> 2.2.1-pyh532a8cf_0
tornado 6.0.3-py37he774522_0 --> 6.0.4-py37he774522_1
vs2015_runtime 14.16.27012-hf0eaf9b_0 --> 14.16.27012-hf0eaf9b_2
wcwidth pkgs/main/win-64::wcwidth-0.1.7-py37_0 --> pkgs/main/noarch::wcwidth-0.2.5-py_0
wheel 0.33.6-py37_0 --> 0.34.2-py37_0
wrapt 1.11.2-py37he774522_0 --> 1.12.1-py37he774522_1
zeromq 4.3.1-h33f27b4_3 --> 4.3.2-ha925a31_2
zipp 0.6.0-py_0 --> 3.1.0-py_0
zlib 1.2.11-h62dcd97_3 --> 1.2.11-h62dcd97_4
| DEBUG menuinst_win32:__init__(199): Menu: name: 'Anaconda${PY_VER} ${PLATFORM}', prefix: 'J:\Anaconda3', env_name: 'None', mode: 'user', used_mode: 'user'
DEBUG menuinst_win32:create(323): Shortcut cmd is %windir%\System32\cmd.exe, args are ['"/K"', 'J:\\Anaconda3\\Scripts\\activate.bat', 'J:\\Anaconda3']
DEBUG menuinst_win32:__init__(199): Menu: name: 'Anaconda${PY_VER} ${PLATFORM}', prefix: 'J:\Anaconda3', env_name: 'None', mode: 'user', used_mode: 'user'
DEBUG menuinst_win32:create(323): Shortcut cmd is %windir%\System32\WindowsPowerShell\v1.0\powershell.exe, args are ['-ExecutionPolicy', 'ByPass', '-NoExit', '-Command', '"& \'J:\\Anaconda3\\shell\\condabin\\conda-hook.ps1\' ; conda activate \'J:\\Anaconda3\' "']
/ DEBUG menuinst_win32:__init__(199): Menu: name: 'Anaconda${PY_VER} ${PLATFORM}', prefix: 'J:\Anaconda3', env_name: 'None', mode: 'user', used_mode: 'user'
DEBUG menuinst_win32:create(323): Shortcut cmd is J:\Anaconda3\pythonw.exe, args are ['J:\\Anaconda3\\cwp.py', 'J:\\Anaconda3', 'J:\\Anaconda3\\pythonw.exe', 'J:\\Anaconda3\\Scripts\\spyder-script.py']
DEBUG menuinst_win32:create(323): Shortcut cmd is J:\Anaconda3\python.exe, args are ['J:\\Anaconda3\\cwp.py', 'J:\\Anaconda3', 'J:\\Anaconda3\\python.exe', 'J:\\Anaconda3\\Scripts\\spyder-script.py', '--reset']
done
C:\WINDOWS\system32>
'3D world > AI Integration Engineer' 카테고리의 다른 글
| Tensorflow, pytorch (0) | 2020.08.10 |
|---|---|
| cat vs dog old ver. & (0) | 2020.07.01 |
| opencv 궁합 맞는 아이들 (0) | 2020.07.01 |
| NVIDIA's GANimation (0) | 2020.07.01 |
| 딥페이크: 인공지능의 혁신과 도전 (0) | 2020.07.01 |
최근댓글