Screen Shot 2018-11-22 at 9.11.45 am

Introducing Highlighter

You may have heard us talk about Silverbrane, a machine learning platform we’ve created over the last two years. We’re very excited to announce that Silverbrane is launching this month, and is now called Highlighter!

Highlighter gives you the freedom to focus on machine learning, rather than the supporting software. We take care of the infrastructure, allowing you to develop and deploy custom machine learning models in an efficient and cost-effective way.

Highlighter provides a seamless interface to label images and train models. These models can then be deployed via an API for inference in production. Progressive improvement of the labelling processes is supported through a reporting and feedback loop.

How does Highlighter work

Upload Data

Upload images that contain objects you want to train a model to identify.

Tag objects of interest

Use the labelling tool to highlight objects of interest in your images.

Watch Highlighter learn

Over time, Highlighter will learn to identify objects for you. The more data you provide, the smarter it gets.

Who is Highlighter for?

Highlighter is for organisations that require a custom machine learning solution, or are interested in developing their own AI capabilities. It is particularly designed for computer vision machine learning projects.

If you already have an AI team, Highlighter enables you to focus on developing your AI capability. If you don’t have an AI team, we’ll guide you through the process of applying the technology to suit your unique requirements.

Machine learning requires a number of steps: preparing data, training a model to identify objects, evaluating the model, making changes, and deploying the algorithm. Highlighter provides the infrastructure for these steps to happen in a cohesive way, so you can easily implement machine learning in your organisation.

Images (Skand 2018): Skand is using Highlighter to reduce the cost and risk attached to asset inspection 

Why use Highlighter?

Highlighter was built by and for machine learning engineers to make computer vision projects simpler. Computer vision is a type of AI that learns to identify objects of interest in images or videos. By automating repetitive visual tasks, Highlighter saves your organisation time and money. It also makes things possible that were previously time-consuming, impractical or difficult for humans to do.

Highlighter supports the full project lifecycle, from data collection to training and production. It supports machine learning labelling, image indexing, activity notification, API integration, reporting and more.

Highlighter is available on the web and via API, and provides users with speed, accuracy, collaboration, and advanced plugin capabilities to tailor to your own workflows. 

Images (Silverpond 2018): The highlighter interface allows your team to easily and efficiently annotate data

Highlighter in Action

While in beta, Highlighter supported several organisations’ machine learning projects with incredible results.

Skand is a Melbourne-based company that reduces the risk and cost of roof inspections. Highlighter formed the basis of their innovative machine-learning approach. By using Highlighter, Skand has:

  • Reduced the cost of roof inspections by 80%
  • Decreased inspection time by 70%
  • Increased coverage of inspections from 15% to 100% of a building’s surface area

 

Wildlife Protection Solutions is on a mission to protect endangered species by reducing poaching. Highlighter was integrated into their process, leading to the detection and capture of poachers at wildlife conservation sites across three continents.

You can read more about Highlighter in action with our Skand case study and Wildlife Protection Solutions case study.

Images (WPS & Silverpond 2018): WPS has been using highlighter to assist conservation rangers catch and detain poachers 

Share this post

Leave a Reply

Your email address will not be published. Required fields are marked *