BUILD & SCALE AI

Advances in deep learning allow for machine vision applications to see more and better than ever before possible. AI can generate competitive advantages when implemented the right way. Use Robovision AI to create machine vision applications.

No coding required to put AI to work

Our software breaks the chains of dependency on data science teams and allows users with basic IT skills to become AI experts in their domain.

Observe, understand and act

Detect defects in your product or classify trash polluting your waterways. The API generated from your pipeline allows you to dictate actions from robots, identify problem areas and suggest potential improvements in real-time.

We push data through pipelines

Chain algorithm cells (such as classifiers, segmentors, (anomaly) detectors, event generators and trackers) and application cells (such as pre-processing, post-processing and normalisation) into intelligent pipelines.

Pipelines used for quality control sometimes need only one perfect example to achieve an accuracy of >99% while others need more labeled examples per output class.

We're highly compatible

  • Reuse existing camera infrastructure, we just need the images
  • Connect to Video Management Systems like Genetec, Milestone or NetworkOptix
  • Our software handles all types of formats of visual data such as (DICOM, point clouds, GIF and more)
  • Through the REST API and OPC broker, our software can be integrated with a wide range of production systems
  • Check our hardware page for the machines that do the thinking and heavy lifting

Designed to scale

Docker is a set of platform as a service products that uses OS-level virtualization to deliver software in packages called containers. Containers are isolated from one another and bundle their own software, libraries and configuration files; they can communicate with each other through well-defined channels

Kubernetes is an open-source system for automating deployment, scaling, and management of containerized applications. It groups containers that make up an application into logical units for easy management and discovery.

Ray is a high-performance distributed execution framework targeted at large-scale machine learning and reinforcement learning applications. It achieves scalability and fault tolerance by abstracting the control state of the system in a global control store and keeping all other components stateless.

Let the algorithms do the hard work

01.

Don’t reinvent the wheel

Robovision AI comes with an AI gallery full of use-case you can start deploying today. There is no need to start from scratch, just connect a use-case pipeline from the AI Gallery to your data and use our teach environment to tweak and fine-tune the models to reach satisfactory accuracy. The AI Gallery can also be deployed in a private environment so you can share intelligence with other departments or entities within your organization.

02.

Don’t do it (by) yourself

Labeling can be one of the most time consuming and annoying tasks to create AI. Our software comes with crowd labeling and predictive labeling functionalities so you don’t have to do it (by) yourself. We create a model automatically based on your first annotations so our predictive labeler can start suggesting labels early on in the labeling process. We also have access to a network of 50,000 labelers to do the labeling for you.

03.

Don’t stop teaching the machines

Deep learning will always work best when trained on your data. There are many variables that impact the performance of your deployment.. With our software you get evaluation capabilities to monitor accuracy, and when suboptimal performance is reported, you can automatically save new data for re-training purposes. In other words, keep fine-tuning and teaching the machines!

Companies that took the head start.

A great run begins with a good start