face-alignment VS DensePose

I used this to detect face landmarks: https://github.com/1adrianb/face-alignment

FROM nvcr.io/nvidia/cuda:10.0-cudnn7-runtime-ubuntu18.04 RUN DEBIAN_FRONTEND=noninteractive apt-get -qq update \ && DEBIAN_FRONTEND=noninteractive apt-get -qqy install python3-pip ffmpeg git less nano libsm6 libxext6 libxrender-dev \ && rm -rf /var/lib/apt/lists/* COPY . /app/ WORKDIR /app RUN pip3 install --upgrade pip RUN pip3 install \ https://download.pytorch.org/whl/cu100/torch-1.0.0-cp36-cp36m-linux_x86_64.whl \ git+https://github.com/1adrianb/face-alignment \ -r requirements.txt ENTRYPOINT [ "python3" ] CMD [ "app.py" ]

Such an annotation format allows the differentiation of face contour, nose, eyes, eyebrows, and lips with a sufficient accuracy level. The data for training the face landmark estimation model might be taken from such open-source libraries as Face Alignment, providing face pose estimation functionality out-of-the-box.

> Densepose: http://densepose.org/

recent example was the team that used Wifi Routers to "See" people They used denspose which is a neural model that can be trained.

Looking at one of the most recent deep learning models DensePose aimed to map pixels of an RGB image of a person to the 3D surface of the human body, we can find out that it’s still not quite suitable for augmented reality. The DensePose’s inference speed is not appropriate for real-time apps, and body mesh detections have insufficient accuracy for the fitting of 3D clothing items. In order to improve results, it’s required to collect more annotated data which is a time and resource-consuming task.