• Stars
    star
    144
  • Rank 255,590 (Top 6 %)
  • Language
    C++
  • License
    GNU General Publi...
  • Created about 7 years ago
  • Updated almost 4 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

(sl)obs filter plug-in for detecting faces and drawing masks on them

obs-facemask-plugin

A libOBS filter plugin that detects faces and draws masks with the detected data.

Compiling

  • Clone this repository with submodules:
	git clone --recursive https://github.com/stream-labs/facemask-plugin
  • Download cmake:

    cmake

  • Get Visual Studio 2015 (vc14) or Visual Studio 2017. When you install it, make sure you include the C++ stuff.

    microsoft

  • Download our fork of OBS Studio:

git clone --recursive https://github.com/stream-labs/obs-studio.git
  • Build obs-studio

    Note: You do not need to install Qt5 as a dependency, just make sure ENABLE_UI in CMake GUI is not checked before Configure/Generate. You can also handle this by setting ENABLE_UI to OFF in CMakeLists in the obs-studio folder.

  • Run cmake in the facemasks folder. When you hit CONFIGURE, you will get errors on fields you need to fill in:

    PATH_OBS_STUDIO Path to the obs-studio folder.

    BUILD_SLOBS - Distributes to slobs instead of OBS Studio

    If you have the Intel Math Kernel Library installed on your system, you might have DLIB_USE_BLAS or DLIB_USE_LAPACK turned on. Keep in mind that dlib links dynamically with these libs, so the MKL and TBB dlls will need to be found by slobs when it runs (for instance, by copying them into the slobs-client folder). I don't reccommend using these libs for this reason.

  • Once you have successfully configured and generated your Visual Studio project with cmake, you can open the facemask-plugin.sln file in Visual Studio. You can now compile the plugin, which will give you a distribution folder structure that mimics the structure in slobs. For example, if you built your files in the build64 folder:

    build64/distribute/slobs/RelWithDebInfo/obs-plugins

    If you are going to debug facemask code make sure to build it as RelWithDebInfo configuration. You can set this in Visual Studio.

  • You can copy the files in manually, or set up symbolic links so you can easily hit F5 and debug from Visual Studio. You can also modify the CMakeLists to copy the files to a desired location with each build.

How It Works

The plugin is an OBS filter plugin. It can be broken down into 2 main parts; the plugin, and the face detection. The face detection runs in its own thread, separate from OBS.

The plugin portion performs the following duties:

  • Render the current frame to a texture
  • Make 3 copies of the current frame, and add them to a circular buffer to be consumed by the face detection thread.
  • Receive new face detection data from another circular buffer fed by the face detection thread.
  • Using smoothing algorithms, the current state of faces is updated.
  • According to the user's parameters, the current faces are rendered.

The face detection portion runs in its own thread. It consumes frame data from the circular buffer, does the face detection computation, then feeds the resulting face data to a circular buffer that is consumed by the rendering.

The process of face detection consists of four main operations:

  • Face Detection The faces are detected using the histogram of oriented gradients method (HOG) in Dlib. HOG's are trained feature descriptors used for detecting objects.

  • Tracking Once we have detected the faces we then use a cheaper method of object tracking to follow the face. This Dlib object tracking method takes an arbitrary rectangle in an image and follows it.

  • Facial Landmarks Given a rectangle that locates a face, we can then use Dlib's landmark detection algorithm which uses a trained regression tree solver to find 68 2D facial landmark points, corresponding to the Multi-PIE definition (see links below).

  • 3D Pose Estimation A subset of key points are taken from the 2D facial landmark points, and using 3D points for an arbitrary rest pose, we use openCV's solvePnP method to obtain a 3D transformation. This transform can be used to render 3D objects in the scene that track the head movement.

  • Face Morphing The 68 landmark points are used to subdivide the video quad into a mesh. Another 11 points are calculated to form the head points, and then catmull rom smoothing is performed to smooth out the contours. Then the mesh is distorted to create face morphs.

The FaceDetect object manages these operations and a current state, so that it performs the face detection, then uses object tracking to follow the face, then does landmark/3d pose estimation, and then the mesh subdivision for face morphing.

Useful Links

Generating thumbnails

Setting up:

  • Install FFmpeg: ffmpeg Add a Path of FFmpeg executable in the Environment Variables: The path should look like: %PATH_TO_YOUR_DIR%\ffmpeg-%VERSION%\bin

  • Install ImageMagick: ImageMagick Turn OFF the Install FFmpeg option during the instalation Turn ON the Install Legacy Utilities (e.g. convert) option during the instalation

  • Download and build giflossy: giflossy Note: Gifview is not needed Add a Path of the gifsicle executable in the Environment Variables: The path should look like: %PATH_TO_YOUR_DIR%\giflossy\src

  • Install SVN: SVN Turn ON the 'Command Line Clients Tool' option during the installation

Generation:

  • Start/Restart Streamlabs-OBS

  • Add a Media Resource

    • Ensure that Loop option is ON and Hide source when playback ends is OFF
    • Choose video by changing properties->local file
  • Add a filter: Face Mask Plugin

  • Go to Setting of facemask-plugin:

    • Set the Demo Mode Folder Path (where the masks JSON files are located)
    • Turn On Generate Thumbs
    • Push Done and wait, thumbnails will be generated in the Demo Folder

Note: if you want to regenerate existing mask, restart the program and remove associated thumbnails from the demo folder

Unit Testing

For Unit Testing CppUtest framework is used

  • During building the facemaks project with CMAKE:
    • Set BUILD_UNIT_TESTS option ON
    • Build and Run the facemask-plugin-test project, it will show the report of tests

Video Comparison side by side

Setting up:

  • Install FFmpeg: ffmpeg Add a Path of FFmpeg executable in the Environment Variables: The path should look like: %PATH_TO_YOUR_DIR%\ffmpeg-%VERSION%\bin

Adding red frame at the start of the video:

Formatting video makes start/end of the looped video detectable by adding a red frame at the start

  • In the facemask-plugin/data there is a script: format.bat
  • Put your sample video as input.mp4 and click on the format.bat
  • Or call the script from the shell
./format.bat /path/to/the/input/video /path/to/the/output/video 

Example:

./format.bat inputfolder/myinput.mp4 outputfolder/myoutput.mp4 

It will generate output with red frame appended

Record Video:

  • Start/Restart Streamlabs-OBS
  • Add a Media Resource
    • Ensure that Loop option is ON and Hide source when playback ends is OFF
    • Choose video by changing properties->local file
There are 2 ways to record video:
  • Automatically detecting start and end using red frame
    This is the best way to use this feature. It will align the two videos perfectly.
    • Add a filter: Face Mask Plugin
    • Go to Setting of facemask-plugin:
      • Set the Demo Mode Folder Path (where the output video should be generated)
    • Tick On Demo Record
      • If it is not working, untick and tick Demo Record again. It will only record the video the first time it is ticked.
    • Recording and Rendering will take some time. The video needs to loop through until red frames are detected twice. Note that the video in the slobs will freeze while the video is being extracted. You just have to wait until it is done.
  • Manually starting and finishing the recording
    • Start Steramlabs-OBS recorder (don't forget to change output format (in settings/output) to mp4)
    • Start and End recording as you see red frame appears

Generate Side By Side Video:

  • Go to Setting of facemask-plugin:
    • Choose the before video
    • Configure the before text for the video
    • Choose the after video
    • Configure the after text for the video
    • Push the Generate button
  • Or call the script from the shell
./sidebyside.bat /path/to/the/beforeVideo /path/to/the/afterVideo before-text after-text /path/to/output/file

More Repositories

1

desktop

Free and open source streaming software built on OBS and Electron.
TypeScript
4,515
star
2

obs-studio-node

libOBS (OBS Studio) for Node.Js, Electron and similar tools
C++
609
star
3

streamlabs-api-demo

Demo of Streamlabs API. Documentation Here https://dev.streamlabs.com
JavaScript
47
star
4

streamlabs-desktop-api-docs

API documentation + web-client example
HTML
43
star
5

lib-streamlabs-ipc

High performance IPC for your native Application
C++
30
star
6

slobs-virtual-cam

Streamlabs Desktop cross-platform virtual camera plugin
C++
28
star
7

streamlabs-overlay

C++
14
star
8

beaker

Vue
14
star
9

platform-example-apps

A collection of example apps for the Streamlabs OBS app platform.
HTML
12
star
10

mediasoup-connector

A libOBS plugin that implements video conferences as sources. https://streamlabs.com/collab-cam
C++
9
star
11

node-window-rendering

Node module to create and render textures on child windows
CMake
8
star
12

a-files-updater

Another updater tool
C++
5
star
13

node-libuiohook

C++
4
star
14

symsrv-scripts

Powershell scripts for Streamlabs Desktop to populate the symbol/source server
PowerShell
4
star
15

crash-handler

Crash handler
C
3
star
16

obs-frontend-api-shim

A simple no-op DLL to fulfill the requirement for the obs-frontend-api library to exist
C++
3
star
17

frontend-assignment

3
star
18

obs-sl-browser

C++
3
star
19

slap

TypeScript
2
star
20

lib-datalane

An incredibly fast IPC solution for C and C++.
C++
2
star
21

color-picker

CMake
1
star
22

streamlabs-desktop-api

Public issue tracking for the Streamlabs Desktop API
1
star
23

lib-streamlabs-datalane

An incredibly fast IPC solution for C and C++.
C++
1
star
24

demo-api-usage

C++
1
star
25

sl-vue-tree

JavaScript
1
star
26

data-engineering-assignment

1
star