JeVois smart machine vision camera

JeVois Smart Machine Vision Camera

Open-source quad-core camera effortlessly adds powerful machine vision

to all your PC, Arduino, and Raspberry Pi projects.

Get one!
Get one!

Introducing JeVois-Pro

Now available for order, ships now! New JeVois-Pro deep learning smart camera packs a 6-core processor and up to 31 trillion operations per seconds (TOPS) of deep neural network processing inside.

JeVois-Pro has an integrated 5-TOPS Neural Processing Unit (NPU), and works with optional 26-TOPS Hailo-8 nueral accelerator, 4-TOPS Google Coral EdgeTPU, and 1-TOPS Intel Myriad-X. Up to 31 TOPS inside (NPU + Hailo-8), and more using additional external USB accelerators.

See JeVois-Pro deep learning benchmarks and JeVois + Hailo-8 demo video.

JeVois is an official partner of Google Coral, bringing you the latest capabilities in embedded deep learning, a solid edge computing development workflow based on Tensorflow-Lite, and amazing support.


About JeVois-A33

Open-source machine vision finally ready for prime-time in all your projects!

JeVois-A33 = video sensor + quad-core CPU + USB video + serial port, all in a tiny, self-contained package (28 cc or 1.7 cubic inches, 17 grams or 0.6 oz). Insert a microSD card loaded with the provided open-source computer vision algorithms (including OpenCV 4.x, TensorFlow, Caffe, Darknet, and many others), connect to your desktop, laptop, and/or Arduino, and give your projects the sense of sight immediately.

Intro video from our January 2017 Kickstarter

1,375 backers pledged $108,397 to help bring this project to life (216% funded).

Do you like buzzwords? JeVois is the world's smallest TensorFlow-enabled deep learning edge AI smart camera for IoT, edge computing, computer vision, home automation and robotics!


 

Need powerful computer vision, artificial intelligence, and a deep learning camera for your project? JeVois runs deep neural networks that can recognize 1000 different types of objects in real time. Watch the video

  Standard Camera:
  • Captures video
  • Sends video to human user
  • Human brain interprets what the video contains

 

 

JeVois Smart Camera:
  • Captures video
  • Processes video to interpret contents
  • Sends interpretation to human, robot, or another computer
Get one!

What can you do with JeVois?

JeVois-Pro
deep neural nets
JeVois-Pro mediapipe
hand detection
JeVois-Pro mediapipe
face mesh detection
JeVois-Pro parallel
DNN accelerators
 
 
Locate and identify
faces or objects
Locate and identify
faces or objects in Python
Recognize 1000 object
types in Python
Classify facial emotions
in Python
 
 
Detect and recognize
1000 object types
Recognize 1000 objects
at up to 83 fps
Detect and recognize
1000 object types
Recognize 1000
objects types
 
 
Locate and identify
objects in scenes
Vision example for
FIRST Robotics 2018
Detect attention-grabbing
objects
Find faces and
handwritten digits
 
 
Detect roads
Detect and record
surprising events
Detect and decode
QR-codes and barcodes
Detect augmented
reality markers
 
 
Detect and decode
ArUco tags
Detect multiple
kinds of tags
Detect objects
by keypoint matching
Track objects
by color
 
 
Compute motion
flow at 100 Hz
Detect and match
salient landmarks
Find
moving objects
Track pupil and
gaze at 120 Hz
 
 
Filter video
using OpenGL-ES 2.0
Morphological
processing
Segment video
into super-pixels
Detect edges at
4 scales simultaneously
 
 
Grab the 3 most
salient regions of interest
Count
dice pips
Fast image
filtering using NEON
Record video
to microSD
 
 
Dense
SIFT keypoints
FIRST Robotics
object detection
ArUco tag +
color blob tracking
Your algorithm
in Python + OpenCV
 
 
Optional global shutter
sensor with IMU
Digital motion processing
on global shutter sensor IMU

 

And more in the full list of bundled machine vision modules -- all are open-source and well documented, to help you learn and develop your own.

Get one!

Getting started with JeVois

Get started on JeVois Start

Try the new JeVois Inventor (beta) graphical frontend for JeVois -- from zero to a working 90 frames/s ArUco tag detector in less than 5 minutes. Full Python 3.7 + OpenCV 4.x support, including deep learning.


Want to discover what JeVois can do? Take the guided tour. Try it with your kids!

Download Guided Tour               
Get started

Quick start, tutorials, downloads and resources

Step-by-step instructions to get you going quickly.

Find out more  Open Access

Vision Modules

Bundled machine vision modules

30+ vision algorithms with full source code to inspire you.

Find out more  Open Source

Intro doc

General introduction and core documentation

From concept to C++17 implementation.

Find out more  Open Source

Programmer Doc

Programmer documentation for vision modules

Open-source, well documented, state-of-the-art -- great for learning.

Find out more  Open Source

Community

Community questions & answers

We are here to help you.

Find out more  Open Access

JeVois on GitHub