Merge 398304f0ca into 080d6f5110

2024-08-13 12:48:38 +00:00 · 2024-08-13 12:48:38 +00:00 · a205feeaba
parent 080d6f5110 398304f0ca
commit a205feeaba
5 changed files with 287 additions and 37 deletions
--- a/README-Fixed.md
+++ b/README-Fixed.md
@ -0,0 +1,139 @@
+**免责声明**
+
+**这个分支经过 NITSC 的修改，只经过我们（NITSC）的电脑验证，没有经过大众验证。**
+这款软件旨在为蓬勃发展的AI生成媒体行业做出积极贡献，帮助艺术家完成动画自定义角色、使用角色作为服装模型等任务。
+开发人员意识到该软件可能存在不道德的应用，并承诺采取预防措施。它内置了检查功能，防止程序在包括裸露、图形内容、战争画面等在内的不适当媒体上运行。我们将继续积极开发该项目，并遵守法律和道德规范。如果法律要求，该项目可能会关闭或在输出中包含水印。
+用户应负责任地使用该软件，并遵守当地法律。如果使用真实人物的面孔，建议用户从相关人员那里获得许可，并在在线发布内容时明确说明这是深度伪造视频。该软件的开发人员不承担最终用户行为的责任。
+
+**如何安装**？
+### 基本安装 (CPU)
+1. **设置平台**:
+    - python (推荐使用 3.10)
+    - pip
+    - git
+    - [ffmpeg](https://www.youtube.com/watch?v=OlNWCpFdVMA) 
+    - [visual studio 2022 runtimes (windows)](https://visualstudio.microsoft.com/visual-cpp-build-tools/)
+    - C:\Users\<用户名>\.keras\keras.json 改为以下内容：
+   ```
+    {
+    "image_data_format": "channels_last",
+    "epsilon": 1e-07,
+    "floatx": "float32",
+    "backend": "tensorflow"
+    }
+   ```
+2. **克隆仓库**:
+    ```
+    https://github.com/hacksider/Deep-Live-Cam.git
+    ```
+3. **下载模型**:
+    1. [GFPGANv1.4](https://huggingface.co/hacksider/deep-live-cam/resolve/main/GFPGANv1.4.pth)
+    2. [inswapper_128_fp16.onnx](https://huggingface.co/hacksider/deep-live-cam/resolve/main/inswapper_128_fp16.onnx)
+    然后将这两个文件放在“**models**”文件夹中。
+4. **安装依赖项**:
+    我们强烈建议使用 `venv` 以避免问题。
+   
+    (适用于 Python3.12.x)
+    ```
+    pip install -r requirements.txt
+    ```
+    (适用于 Python3.10.x)
+    ```
+    pip install -r orginal_requirements.txt
+    ```
+    完成 !!! 如果您没有 GPU，您应该能够使用 `python run.py` 命令运行 roop。请注意，在首次运行程序时，它将下载一些模型，这可能会根据您的网络连接花费一些时间。
+### GPU 加速
+#### CUDA 执行提供程序 (Nvidia)
+1.  安装 [CUDA Toolkit 11.8](https://developer.nvidia.com/cuda-11-8-0-download-archive)
+    
+2.  安装依赖项:
+    ```
+    pip uninstall onnxruntime onnxruntime-gpu
+    pip install onnxruntime-gpu==1.16.3
+    ```
+3.  如果提供程序可用，则使用:
+    ```
+    python run.py --execution-provider cuda
+    ```
+#### CoreML 执行提供程序 (Apple Silicon)
+1.  安装依赖项:
+    ```
+    pip uninstall onnxruntime onnxruntime-silicon
+    pip install onnxruntime-silicon==1.13.1
+    ```
+2.  如果提供程序可用，则使用:
+    ```
+    python run.py --execution-provider coreml
+    ```
+#### CoreML 执行提供程序 (Apple Legacy)
+1.  安装依赖项:
+    ```
+    pip uninstall onnxruntime onnxruntime-coreml
+    pip install onnxruntime-coreml==1.13.1
+    ```
+2.  如果提供程序可用，则使用:
+    ```
+    python run.py --execution-provider coreml
+    ```
+#### DirectML 执行提供程序 (Windows)
+1.  安装依赖项:
+    ```
+    pip uninstall onnxruntime onnxruntime-directml
+    pip install onnxruntime-directml==1.15.1
+    ```
+2.  如果提供程序可用，则使用:
+    ```
+    python run.py --execution-provider dml
+    ```
+#### OpenVINO™ 执行提供程序 (Intel)
+1.  安装依赖项:
+    ```
+    pip uninstall onnxruntime onnxruntime-openvino
+    pip install onnxruntime-openvino==1.15.0
+    ```
+2.  如果提供程序可用，则使用:
+    ```
+    python run.py --execution-provider openvino
+    ```
+**如何使用**？
+> 注意：首次运行此程序时，它将下载一些模型，大小约为 300MB。
+执行 `python run.py` 命令将启动以下窗口：
+![gui-demo](instruction.png)
+选择一个面部 (包含所需面部的图像) 和目标图像/视频 (您想要替换面部的图像/视频)，然后单击“开始”。打开文件资源管理器，导航到您选择的输出目录。您将找到名为 `<video_title>` 的目录，其中可以实时查看交换的帧。处理完成后，它将创建输出文件。就这样。
+**网络摄像头模式**
+只需按照屏幕截图上的步骤操作：
+1. 选择一个面部
+2. 单击“实时”
+3. 等待几秒钟 (通常需要 10 到 30 秒才能显示预览)
+![demo-gif](demo.gif)
+只需使用您喜欢的屏幕录制软件进行直播，例如 OBS。
+> 注意：如果您想更改您的面部，只需选择另一张图片，预览模式将重新启动 (所以只需等待一会儿)。
+以下是一些额外的命令行参数。要了解它们的功能，请查看 [此指南](https://github.com/s0md3v/roop/wiki/Advanced-Options)。
+```
+options:
+  -h, --help                                               显示此帮助消息并退出
+  -s SOURCE_PATH, --source SOURCE_PATH                     选择源图像
+  -t TARGET_PATH, --target TARGET_PATH                     选择目标图像或视频
+  -o OUTPUT_PATH, --output OUTPUT_PATH                     选择输出文件或目录
+  --frame-processor FRAME_PROCESSOR [FRAME_PROCESSOR ...]  帧处理器 (choices: face_swapper, face_enhancer, ...)
+  --keep-fps                                               保持原始 fps
+  --keep-audio                                             保持原始音频
+  --keep-frames                                            保留临时帧
+  --many-faces                                             处理每个面部
+  --video-encoder {libx264,libx265,libvpx-vp9}             调整输出视频编码器
+  --video-quality [0-51]                                   调整输出视频质量
+  --max-memory MAX_MEMORY                                  最大 RAM 量 (GB)
+  --execution-provider {cpu} [{cpu} ...]                   可用的执行提供程序 (choices: cpu, ...)
+  --execution-threads EXECUTION_THREADS                    执行线程数
+  -v, --version                                            显示程序的版本号并退出
+```
+想要 CLI 模式？使用 -s/--source 参数将使 run 程序以 CLI 模式运行。
+**想要立即获得下一个更新**？
+如果您想要最新的构建版本或想体验一些新的功能，请转到我们的 [experimental branch](https://github.com/hacksider/Deep-Live-Cam/tree/experimental) 并体验贡献者带来的功能。
+**致谢**
+- [ffmpeg](https://ffmpeg.org/): 让视频相关操作变得容易
+- [deepinsight](https://github.com/deepinsight): 他们的 [insightface](https://github.com/deepinsight/insightface) 项目提供了一个制作精良的库和模型。
+- [havok2-htwo](https://github.com/havok2-htwo) : 分享用于网络摄像头的代码
+- [GosuDRM](https://github.com/GosuDRM/nsfw-roop) : 解除 roop 的审查
+- 以及 [所有开发者](https://github.com/hacksider/Deep-Live-Cam/graphs/contributors) 在该项目中使用的库背后的所有开发者。
+- 脚注：[这原本是 roop-cam，请在此处查看代码的完整历史。](https://github.com/hacksider/roop-cam) 请注意，代码的基础作者是 [s0md3v](https://github.com/s0md3v/roop)
--- a/modules/ui.py
+++ b/modules/ui.py
@ -61,47 +61,47 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
    target_label = ctk.CTkLabel(root, text=None)
    target_label.place(relx=0.6, rely=0.1, relwidth=0.3, relheight=0.25)

-    source_button = ctk.CTkButton(root, text='Select a face', cursor='hand2', command=lambda: select_source_path())
+    source_button = ctk.CTkButton(root, text='选择一个面部', cursor='hand2', command=lambda: select_source_path())
    source_button.place(relx=0.1, rely=0.4, relwidth=0.3, relheight=0.1)

-    target_button = ctk.CTkButton(root, text='Select a target', cursor='hand2', command=lambda: select_target_path())
+    target_button = ctk.CTkButton(root, text='选择一个目标', cursor='hand2', command=lambda: select_target_path())
    target_button.place(relx=0.6, rely=0.4, relwidth=0.3, relheight=0.1)

    keep_fps_value = ctk.BooleanVar(value=modules.globals.keep_fps)
-    keep_fps_checkbox = ctk.CTkSwitch(root, text='Keep fps', variable=keep_fps_value, cursor='hand2', command=lambda: setattr(modules.globals, 'keep_fps', not modules.globals.keep_fps))
+    keep_fps_checkbox = ctk.CTkSwitch(root, text='保持帧率', variable=keep_fps_value, cursor='hand2', command=lambda: setattr(modules.globals, 'keep_fps', not modules.globals.keep_fps))
    keep_fps_checkbox.place(relx=0.1, rely=0.6)

    keep_frames_value = ctk.BooleanVar(value=modules.globals.keep_frames)
-    keep_frames_switch = ctk.CTkSwitch(root, text='Keep frames', variable=keep_frames_value, cursor='hand2', command=lambda: setattr(modules.globals, 'keep_frames', keep_frames_value.get()))
+    keep_frames_switch = ctk.CTkSwitch(root, text='保留镜框', variable=keep_frames_value, cursor='hand2', command=lambda: setattr(modules.globals, 'keep_frames', keep_frames_value.get()))
    keep_frames_switch.place(relx=0.1, rely=0.65)

    # for FRAME PROCESSOR ENHANCER tumbler:
    enhancer_value = ctk.BooleanVar(value=modules.globals.fp_ui['face_enhancer'])
-    enhancer_switch = ctk.CTkSwitch(root, text='Face Enhancer', variable=enhancer_value, cursor='hand2', command=lambda: update_tumbler('face_enhancer',enhancer_value.get()))
+    enhancer_switch = ctk.CTkSwitch(root, text='面部优化', variable=enhancer_value, cursor='hand2', command=lambda: update_tumbler('face_enhancer',enhancer_value.get()))
    enhancer_switch.place(relx=0.1, rely=0.7)

    keep_audio_value = ctk.BooleanVar(value=modules.globals.keep_audio)
-    keep_audio_switch = ctk.CTkSwitch(root, text='Keep audio', variable=keep_audio_value, cursor='hand2', command=lambda: setattr(modules.globals, 'keep_audio', keep_audio_value.get()))
+    keep_audio_switch = ctk.CTkSwitch(root, text='保留音频', variable=keep_audio_value, cursor='hand2', command=lambda: setattr(modules.globals, 'keep_audio', keep_audio_value.get()))
    keep_audio_switch.place(relx=0.6, rely=0.6)

    many_faces_value = ctk.BooleanVar(value=modules.globals.many_faces)
-    many_faces_switch = ctk.CTkSwitch(root, text='Many faces', variable=many_faces_value, cursor='hand2', command=lambda: setattr(modules.globals, 'many_faces', many_faces_value.get()))
+    many_faces_switch = ctk.CTkSwitch(root, text='多脸', variable=many_faces_value, cursor='hand2', command=lambda: setattr(modules.globals, 'many_faces', many_faces_value.get()))
    many_faces_switch.place(relx=0.6, rely=0.65)

    nsfw_value = ctk.BooleanVar(value=modules.globals.nsfw)
    nsfw_switch = ctk.CTkSwitch(root, text='NSFW', variable=nsfw_value, cursor='hand2', command=lambda: setattr(modules.globals, 'nsfw', nsfw_value.get()))
    nsfw_switch.place(relx=0.6, rely=0.7)

-    start_button = ctk.CTkButton(root, text='Start', cursor='hand2', command=lambda: select_output_path(start))
+    start_button = ctk.CTkButton(root, text='开始', cursor='hand2', command=lambda: select_output_path(start))
    start_button.place(relx=0.15, rely=0.80, relwidth=0.2, relheight=0.05)

-    stop_button = ctk.CTkButton(root, text='Destroy', cursor='hand2', command=lambda: destroy())
+    stop_button = ctk.CTkButton(root, text='关闭', cursor='hand2', command=lambda: destroy())
    stop_button.place(relx=0.4, rely=0.80, relwidth=0.2, relheight=0.05)

-    preview_button = ctk.CTkButton(root, text='Preview', cursor='hand2', command=lambda: toggle_preview())
+    preview_button = ctk.CTkButton(root, text='预览', cursor='hand2', command=lambda: toggle_preview())
    preview_button.place(relx=0.65, rely=0.80, relwidth=0.2, relheight=0.05)

-    live_button = ctk.CTkButton(root, text='Live', cursor='hand2', command=lambda: webcam_preview())
+    live_button = ctk.CTkButton(root, text='实时', cursor='hand2', command=lambda: webcam_preview())
    live_button.place(relx=0.40, rely=0.86, relwidth=0.2, relheight=0.05)

    status_label = ctk.CTkLabel(root, text=None, justify='center')
@ -120,7 +120,7 @@ def create_preview(parent: ctk.CTkToplevel) -> ctk.CTkToplevel:

    preview = ctk.CTkToplevel(parent)
    preview.withdraw()
-    preview.title('Preview')
+    preview.title('预览')
    preview.configure()
    preview.protocol('WM_DELETE_WINDOW', lambda: toggle_preview())
    preview.resizable(width=False, height=False)
@ -146,7 +146,7 @@ def select_source_path() -> None:
    global RECENT_DIRECTORY_SOURCE, img_ft, vid_ft

    PREVIEW.withdraw()
-    source_path = ctk.filedialog.askopenfilename(title='select an source image', initialdir=RECENT_DIRECTORY_SOURCE, filetypes=[img_ft])
+    source_path = ctk.filedialog.askopenfilename(title='选择源图像', initialdir=RECENT_DIRECTORY_SOURCE, filetypes=[img_ft])
    if is_image(source_path):
        modules.globals.source_path = source_path
        RECENT_DIRECTORY_SOURCE = os.path.dirname(modules.globals.source_path)
@ -161,7 +161,7 @@ def select_target_path() -> None:
    global RECENT_DIRECTORY_TARGET, img_ft, vid_ft

    PREVIEW.withdraw()
-    target_path = ctk.filedialog.askopenfilename(title='select an target image or video', initialdir=RECENT_DIRECTORY_TARGET, filetypes=[img_ft, vid_ft])
+    target_path = ctk.filedialog.askopenfilename(title='选择目标图像或视频', initialdir=RECENT_DIRECTORY_TARGET, filetypes=[img_ft, vid_ft])
    if is_image(target_path):
        modules.globals.target_path = target_path
        RECENT_DIRECTORY_TARGET = os.path.dirname(modules.globals.target_path)
@ -181,9 +181,9 @@ def select_output_path(start: Callable[[], None]) -> None:
    global RECENT_DIRECTORY_OUTPUT, img_ft, vid_ft

    if is_image(modules.globals.target_path):
-        output_path = ctk.filedialog.asksaveasfilename(title='save image output file', filetypes=[img_ft], defaultextension='.png', initialfile='output.png', initialdir=RECENT_DIRECTORY_OUTPUT)
+        output_path = ctk.filedialog.asksaveasfilename(title='保存输出图像', filetypes=[img_ft], defaultextension='.png', initialfile='output.png', initialdir=RECENT_DIRECTORY_OUTPUT)
    elif is_video(modules.globals.target_path):
-        output_path = ctk.filedialog.asksaveasfilename(title='save video output file', filetypes=[vid_ft], defaultextension='.mp4', initialfile='output.mp4', initialdir=RECENT_DIRECTORY_OUTPUT)
+        output_path = ctk.filedialog.asksaveasfilename(title='保存输出视频', filetypes=[vid_ft], defaultextension='.mp4', initialfile='output.mp4', initialdir=RECENT_DIRECTORY_OUTPUT)
    else:
        output_path = None
    if output_path:
--- a/orginal_requirements.txt
+++ b/orginal_requirements.txt
@ -0,0 +1,23 @@
+--extra-index-url https://download.pytorch.org/whl/cu118
+
+numpy==1.23.5
+opencv-python==4.8.1.78
+onnx==1.16.0
+insightface==0.7.3
+psutil==5.9.8
+tk==0.1.0
+customtkinter==5.2.2
+pillow==9.5.0
+torch==2.0.1+cu118; sys_platform != 'darwin'
+torch==2.0.1; sys_platform == 'darwin'
+torchvision==0.15.2+cu118; sys_platform != 'darwin'
+torchvision==0.15.2; sys_platform == 'darwin'
+onnxruntime==1.18.0; sys_platform == 'darwin' and platform_machine != 'arm64'
+onnxruntime-silicon==1.16.3; sys_platform == 'darwin' and platform_machine == 'arm64'
+onnxruntime-gpu==1.18.0; sys_platform != 'darwin'
+tensorflow==2.13.0rc1; sys_platform == 'darwin'
+tensorflow==2.12.1; sys_platform != 'darwin'
+opennsfw2==0.10.2
+protobuf==4.23.2
+tqdm==4.66.4
+gfpgan==1.3.8
--- a/requirements.txt
+++ b/requirements.txt
@ -1,23 +1,108 @@
--extra-index-url https://download.pytorch.org/whl/cu118
-
-numpy==1.23.5
-opencv-python==4.8.1.78
-onnx==1.16.0
-insightface==0.7.3
-psutil==5.9.8
-tk==0.1.0
+absl-py==2.1.0
+addict==2.4.0
+albucore==0.0.13
+albumentations==1.4.13
+annotated-types==0.7.0
+astunparse==1.6.3
+basicsr==1.4.2
+beautifulsoup4==4.12.3
+certifi==2024.7.4
+charset-normalizer==3.3.2
+colorama==0.4.6
+coloredlogs==15.0.1
+contourpy==1.2.1
 customtkinter==5.2.2
-pillow==9.5.0
-torch==2.0.1+cu118; sys_platform != 'darwin'
-torch==2.0.1; sys_platform == 'darwin'
-torchvision==0.15.2+cu118; sys_platform != 'darwin'
-torchvision==0.15.2; sys_platform == 'darwin'
-onnxruntime==1.18.0; sys_platform == 'darwin' and platform_machine != 'arm64'
-onnxruntime-silicon==1.16.3; sys_platform == 'darwin' and platform_machine == 'arm64'
-onnxruntime-gpu==1.18.0; sys_platform != 'darwin'
-tensorflow==2.13.0rc1; sys_platform == 'darwin'
-tensorflow==2.12.1; sys_platform != 'darwin'
-opennsfw2==0.10.2
-protobuf==4.23.2
-tqdm==4.66.4
+cycler==0.12.1
+Cython==3.0.11
+darkdetect==0.8.0
+easydict==1.13
+eval_type_backport==0.2.0
+facexlib==0.3.0
+filelock==3.15.4
+filterpy==1.4.5
+flatbuffers==24.3.25
+fonttools==4.53.1
+fsspec==2024.6.1
+future==1.0.0
+gast==0.6.0
+gdown==5.2.0
 gfpgan==1.3.8
+google-pasta==0.2.0
+grpcio==1.65.4
+h5py==3.11.0
+humanfriendly==10.0
+idna==3.7
+imageio==2.34.2
+importlib_metadata==8.2.0
+insightface==0.7.3
+Jinja2==3.1.4
+joblib==1.4.2
+keras==3.5.0
+kiwisolver==1.4.5
+lazy_loader==0.4
+libclang==18.1.1
+llvmlite==0.43.0
+lmdb==1.5.1
+Markdown==3.6
+markdown-it-py==3.0.0
+MarkupSafe==2.1.5
+matplotlib==3.9.1.post1
+mdurl==0.1.2
+ml-dtypes==0.4.0
+mpmath==1.3.0
+namex==0.0.8
+networkx==3.3
+numba==0.60.0
+numpy==1.26.4
+onnx==1.16.0
+onnxruntime==1.18.0
+opencv-python==4.8.1.78
+opencv-python-headless==4.10.0.84
+opennsfw2==0.10.2
+opt-einsum==3.3.0
+optree==0.12.1
+packaging==24.1
+Pillow==9.5.0
+platformdirs==4.2.2
+prettytable==3.11.0
+protobuf==4.23.2
+psutil==5.9.8
+pydantic==2.8.2
+pydantic_core==2.20.1
+Pygments==2.18.0
+pyparsing==3.1.2
+pyreadline3==3.4.1
+PySocks==1.7.1
+python-dateutil==2.9.0.post0
+PyYAML==6.0.2
+realesrgan==0.3.0
+requests==2.32.3
+rich==13.7.1
+scikit-image==0.24.0
+scikit-learn==1.5.1
+scipy==1.14.0
+setuptools==72.1.0
+six==1.16.0
+soupsieve==2.5
+sympy==1.13.2
+tb-nightly==2.18.0a20240812
+tensorboard==2.17.0
+tensorboard-data-server==0.7.2
+tensorflow==2.17.0
+tensorflow-intel==2.17.0
+termcolor==2.4.0
+threadpoolctl==3.5.0
+tifffile==2024.8.10
+tk==0.1.0
+tomli==2.0.1
+torch==2.2.0
+torchvision==0.17.0
+tqdm==4.66.4
+typing_extensions==4.12.2
+urllib3==2.2.2
+wcwidth==0.2.13
+Werkzeug==3.0.3
+wheel==0.44.0
+wrapt==1.16.0
+yapf==0.40.2
+zipp==3.20.0
--- a/run.py
+++ b/run.py
@ -2,5 +2,8 @@

 from modules import core

+import os
+os.environ['KERAS_BACKEND'] = 'tensorflow'
+
 if __name__ == '__main__':
    core.run()