Skip to content

Latest commit

 

History

History
68 lines (52 loc) · 1.98 KB

CONTRIBUTING.md

File metadata and controls

68 lines (52 loc) · 1.98 KB
Raw

Contribution Guidelines

How to contribution

🖐 Make issue

  1. Prepare demo gif and link for your mobile machine learning project.

I recommend the width of gif is 320.
If you want to know how to generate gif demo, check this tutorial.

  1. Make an issue before PR

I will prepare the issue guidelines, but not yet.

👍 Make PR

  1. Fork Awesome Machine Learning DEMOs with iOS repository.
  2. Add your demo gif and link.
  3. Make PR.

Tips

How to generate gif

  1. Record DEMO video on iOS.
    → How to record the screen on your iPhone, iPad, or iPod touch
  2. Export the video to mac.
  3. Convert the video to gif using following script.
  4. Install FFMPEG on your mac.
    → Run the following command on Terminal app.
    $ brew install ffmpeg
  5. Download the script(gifconverter.sh).
# echo $1
video_path=$1
directory_path=`dirname "$1"`
file=`basename "$1"`

extension="${file##*.}"
filename="${file%.*}"
echo $extension

gif_path="${video_path%.*}.gif"
echo $gif_path

ffmpeg  -y -i $video_path  -vf fps=8,scale=320:-1:flags=lanczos,palettegen palette.png;
ffmpeg -y -i $video_path -i palette.png -filter_complex "fps=8,scale=320:-1:flags=lanczos[x];[x][1:v]paletteuse" $gif_path;
  1. Run the script with the demo video path.
    $ sh gifconverter.sh ~/Desktop/DEMO.mp4
  2. You can find DEMO.gif on ~/Desktop.

How to measure the performance

Unit Test

Show output for each input? Drawing detail of result? Test for debugging?

  • Pose Estimation: draw dot each point and joint, print confidence each point.
  • ...

Bunch Test (planning...)

Analyze outputs from a bunch of inputs

  • average of inference time and fps
  • accumulate execution time, fps...?
  • rendering time
  • total execution time
  • ...