Build a face recognition app with AI edge computing
People detection and tracking
Object tracking is a computer vision task used to follow the movement of objects across video
frames, helping systems monitor and understand how things change over time. It is used in
a wide range of real-world scenarios from estimating speed and direction of moving objects to
counting objects in videos.
Systems integrated with object tracking can be configured to trigger real-time alerts when
something unusual is detected, such as a person entering a restricted area or a delivery being
left too long in one place.
Lightweight face recognition
Detect and identify multiple faces directly from your browser or use the
Abraia Vision SDK to develop your custom
application. Simply take a picture and label the people you want to recognize and test the face
recognition solution. Ready to deploy on your edge hardware using the onnxruntime.
Recognize faces directly on your hardware avoiding privacy concerns.
How to build a real-time face recognition tracking application
Object detection is a computer vision task that involves identifying and locating objects within an
image or video. It has numerous applications, including surveillance, autonomous vehicles, and robotics.
In this tutorial, we'll show you how easy is identifying people on image or video with the
Abraia Vision SDK. Just
follow these simple steps:
1. Install the Abraia Python SDK
You can install the package on Windows, Mac, and Linux:
python -m pip install -U abraia
2. Load and run the face recognition model
Import the FaceRecognizer class from abraia inference. Then use "load_image" and run the model to
detect every face in the image. Build the index with a portrait image of every person you want to
identify in the image. Finally, identify people on the image using the previously generated search
index, and use "render_results" to show the results on the image.
import os
from abraia.inference import FaceRecognizer
from abraia.utils import load_image, save_image, render_results
img = load_image('images/rolling-stones.jpg')
out = img.copy()
recognition = FaceRecognizer()
results = recognition.represent_faces(img)
index = []
for src in ['mick-jagger.jpg', 'keith-richards.jpg', 'ronnie-wood.jpg', 'charlie-watts.jpg']:
img = load_image(f"images/{src}")
rslt = recognition.represent_faces(img)[0]
index.append({'name': os.path.splitext(src)[0], 'embeddings': rslt['embeddings']})
results = recognition.identify_faces(results, index)
render_results(out, results)
save_image(out, 'images/rolling-stones-identified.jpg')
3. Run the face recognition model on video
Import the class "Video" to create a video object to iterate on the frames. Input the path
for a video file or "Video(0)" to use your webcam. So, you can directly run the model directly
on your camera stream, and show the results using the "show" method from video.
from abraia.utils import Video
from abraia.inference import Tracker
video = Video(0)
tracker = Tracker(frame_rate=video.frame_rate)
for frame in video:
results = recognition.represent_faces(frame)
results = recognition.identify_faces(results, index)
results = tracker.update(results)
frame = render_results(frame, results)
video.show(frame)
Congratulations! You have successfully run the face recognition model on a video. This tutorial provides
a basic overview, and you can further explore advanced features, optimize your model, and fine-tune
parameters based on your specific use case.
