Ikomia : the best Computer Vision algorithms in a few lines of code

Published by Guillaume Demarcq on

Ikomia – A shared vision of research

Ikomia is a startup dedicated to computer vision and artificial intelligence.

At Ikomia, we deeply believe in a modern way to use and share research works all around the world.

We also believe that accessing and testing methods from the state of the art should not be time consuming.

Finally, we believe that researchers make some promising algorithms but unfortunately, most of them are not exploited by the industry. Hence, we want to simplify all these steps and we propose this exclusive solution to ease your computer vision experience.

We created the open source Ikomia platform as a universal and optimized platform for using and sharing algorithms. Let’s take a tour.

Input data

 

Input data

The platform uses OpenCV and VTK to load the following data:

  • Color and grayscale images in standard formats
  • Videos
  • Temporal image sequences
  • Image stacks (3D)
  • Medical images (DICOM)
  • Streaming (Internal webcam, USB Cam, IP Cam)

Computer vision libraries

Computer vision libIn Ikomia, we offer many algorithms provided by the OpenCV framework. It ranges from the basic operations to some more sophisticated processes. Moreover, we begin to integrate the G’MIC library developed by the GREYC lab. We could integrate many other libraries like Dlib, ITK or VTK. In short, we are waiting for your feedback!

Node based editor

Ikomia vision - Node based editor

 

To simplify the use of many algorithms, we propose Ikomia STUDIO, a node based editor that chains multiple processes. This graph-based approach allows you to design your own algorithm or to compare different methods. Then the limit relies only on your creativity!

 

Vision algorithm integration

Algorithm integration

Do you develop innovative, fast or original algorithms? Do you code in Python or C++?  Then Ikomia API is made for you! Don’t wait and use our API to easily bind your code in the platform! What’s the advantage? You benefit from an already dedicated environment for your data and you can compare your work with others. Moreover, you can promote and share your algorithm with the community thanks to our HUB.

Computer vision HUB

Algorithm marketplace

 

Finally, we built a hub for computer vision and deep learning algorithms. As a user, you can download a tested and ready to use method that you can apply on your own data. As a researcher, you can publish your own ready to use Python code to the community.

Our n°1 goal : building a CV community

After 2 years of hard work and tens of thousands of lines of code, we are eager to get as many feedbacks from our users as possible. If you wish to contribute, you are welcome to do so in the comments below, in the Q&A page of the platform, on Github, LinkedIn or Twitter.

While the team keeps on building new features for CV developers, we are now focus ion building the most collective, sharing and innovative CV community. To join us please register.

 


Follow us on Twitter and Linkedin.

 


0 Comments

Leave a Reply

Avatar placeholder

Your email address will not be published.