Why Scanner?

Standard library of DNN modules

Face Detection

...

Pose Detection

...

Object Detection

...

Simple Python API

from scannertools import face_detection, vis, sample_video
import scannerpy

def main():
    with sample_video() as video: # get an example video
        db = scannerpy.Database() # start up Scanner
        frames = list(range(50))  # select the first 50 frames
        bboxes = face_detection.detect_faces(
            db, videos=[video], frames=[frames])
        vis.draw_bboxes(
            db, videos=[video], frames=[frames],
            bboxes=bboxes, paths=['sample_faces.mp4'])
        # output video is saved to 'sample_faces.mp4'

if __name__ == "__main__":
    main()
Dive into this example

Designed for Performance

1. Never convert your videos into images again

Scanner provides native compressed video support, which enables directly accessing compressed videos to keep your storage and bandwidth costs low.

2. First-class support for GPUs

Scanner supports machines with CPUs and GPUs out of the box so that you can make the most out of your hardware.

3. Cloud & cluster ready

Scanner is integrated with Kubernetes, Google Cloud Platform, and AWS to make it easy to spin up machines to get your results back faster.

Scanner in Production

Analyzing the Internet Archive's TV News Database

Use modern neural networks to automatically detect objects, people, and faces in your videos. Check out scannertools for an easy to use toolkit that's setup to download and run these models.

Facebook's Manifold VR Camera

Scanner is currently being used as the compute engine behind the Manifold 360 video camera from Facebook and RED. Scanner has also been integrated into the open-source version of the previous generation Surround 360 system on GitHub.