Changing the paradigm
of computer vision

Embedded computer vision that is fast, cost-effective and precise

A new era established...

Traditional methods of computer vision consume lots of bandwidth or compute resources before even talking about the privacy issues involved. Today, Datakalab ushers in a new era of vision that is anonymous by design, lightweight and precise. Gone are the days of privacy concerns and expensive cloud and GPU solutions. 
Say hello to performance, privacy and low cost CPUs.

Illustration

Private by Design

Images are processed in real time and never stored locally or centrally. The only output from our software are anonymous statistics.

100% on the edge

Datakalab's algorithms were built from the ground up to run locally on x86 CPUs so that video feeds can be processed in real time and independent of any internet connection.

Self Adapting

Datakalab's algorithms automatically take into account the differences in camera lighting conditions that exist or may change over time without needing any additional human data annotation.

Energy and Cost Efficient

No costly cloud processing or bandwidth is needed. GPUs? We don't need no GPUs. Run Datakalab effortlessly on an x86 architecture with pre-existing cameras.

Highly Precise

Datakalab's algorithms track and aggregate data over multiple frames to so that you get the statistics you need with over 97% precision in certain use cases.

Nimble and Fast

With some of the industry's lightest minimum requirements, install Datakalab on your pre-existing hardware infrastructure. Gone are they days of expensive 3D cameras and GPUs.

Illustration

Unicity Detection

Anti Tailgating monitoring 24x7 without expensive 3D cameras or GPUs

  • Gate Manufacturers

    Whether it be at the immigration border, a metro station or a secure entrance for a data center or warehouse, Datakalab's unique unicity detection algorithm helps keep people moving while ensuring no one is tail gating behind. By using vision instead of time of flight, we can see things that others can't. Higher precision while reducing hardware implementation spend.

  • Security System Integrators

    As a system integrator, are your clients looking for innovative solutions to complex problems while respecting an individual's right to privacy? Do you need a lightweight vision solution that can integrate in your client's hypervisor or VMS? Look no further. 

Video Analytics 

Bringing Google Analytics like statistics to the physical world

  • Digital out of Home

    Gain better insight into which advertising placements generate the highest reach. Open new revenue streams by providing people analytics to municipalities.

  • Transportation

    Calculate fraud vs paid users . Let your users check the people density in buses and metros in real time. Determine if you need to put another bus or train in operation.

  • Retail

    Gain insight into the peak traffic times and demographics of your shoppers using pre-existing security cameras. Refine the storefront rent based on people data.

  • Smart Cities

    Learn how people go about your city to better serve them. Get a breakdown of the traffic type (people, cars, cyclists...) per street. 

Video analytics

Success Stories

Heading photo

Improve border crossing

Illustration

Saved over 3k€ per gate in hardware costs and energy consumption while increasing unicity detection precision by 30%.

  • "Datakalab has the advantage of being scalable and allowing IN Groupe to be flexible in taking into account any new requirements or modifications requested by countries or operators and solving the constraints related to the acquisition of expensive edge-based hardware."

    Yann Haguet, EVP at IN Groupe.

Heading photo

Evaluate the ROI

Illustration

Provided metrics to value a DooH panel network through analytics on Audience Volume / Audience Qualification / Audience Consistency.

  • "ExterionMedia's new DOOH digital offering, launched in 2019, quickly established itself as leader on the market, thanks to Datakalab technology, which is able to quantify and qualify traffic flows and audiences for each digital display in real time. Thanks to the embedded audience intelligence that Datakalab offers, ExterionMedia's marketing teams have become one of the main market references in 2021."

    Jean-François Curtil, CEO at ExterionMedia France.

Heading photo

Reinforcing Mask Wearing

Illustration

Allow the broadcast of audio announcements encouraging travelers to wear masks when mask wearing drops below 80%.

  • "Smart cameras can now be installed in public transportation networks allowing operators the ability to measure the percentage of people wearing masks, without using any personal data. I salute Datakalab, a French start-up, helping us maintain social distancing best practices."

    Jean-Baptiste Djebbari, French Minister of Transport.

Doing More With Less

Datakalab specialises in low power, runtime efficient and privacy compliant deep learning algorithms that work entirely on the edge

Illustration

Speed and precision of the future... today

Cutting edge R&D is our DNA. With over 9 research papers and 3 patents pending, our scientists and researchers work non-stop to push the boundaries in lossless AI acceleration and compression.  

Check out our latest papers accepted at the leading International AI conferences NeurIPS, CVPR, ICCV and AAAI and above all. And if you think you can help us push the boundaries even further, hit us up on our career page! 

Latest News

Confiance.ai reveals the 12 winning start-ups and innovative SMEs from its call for expressions of interest

Datakalab is selected to participate in a french community to design and industrialise trustworthy AI-based critical systems.

the-lancet-infectious-diseases

Datakalab demonstrated that 91% of participants wore their masks throughout the concert organised by AP-HP and PRODISS.

Award "Best Service Provider"

Datakalab and IN Groupe present their partnership to meet the requirements of the States and promote a french security system solution.

Have questions?

Tell us what's on your mind. We'd love to talk.