Dear our customers,

Today 20th Sep 2021, we are thrilled to introduce our newest product BlueEye Data to the community.

Video and image manual labeling requires a lot of human resources and can be a very painful process. BlueEye Data is created to solve this problem for you.

1. Core Concept

So, what is BlueEye Data?

BlueEye Data is a leading AI data platform specialized in data labelling that provides high-quality training data and service for AI models/applications to speed up the AI developing process. By providing high-quality training data via smart labelling tools, real-time and innovative collaborations, configurable workflows and domain expert annotators, BlueEye is able to shorten the developing time of an AI application considerably. At the same time, BlueEye Data provides an infrastructure to manage and keep track of any project progress, aiming to bring out the most effective workflow and leading your company to success.

What does BlueEye Data offer?

For business:

High-quality training data for reliable models.
Annotation tools for video, images, text and document.
Independent labeling system that allows users to interact in parallel on many different projects.
A complete system for your organization to collaborate on creating, managing and reviewing training data; instantly track project updates, task status with a notification system.
Credible labeling service to minimize the labeling cost with automatic annotation.

For direct users:
BlueEye Data has four main direct users which are the labelers, the reviewers (ML engineers, Data Scientists), the project managers and the Admins (CTO/CEO)

Labelers: BlueEye Data makes the labeling process much easier and increases productivity by providing smart tools (both human-powered and AI assisted), easy collaborative workflow, modern and easy-to-use interface.
Reviewers: easily to get insights, review and validate training data; results in a better and faster ML model building process.
Project Managers and Admins: transparency and good visibility of AI projects from clear dashboards and KPI/metrics

Main features

Manage projects: BlueEye Data has a comprehensive management platform to set up new projects, overview current running projects, and manage involved members. The product is able to track multiple projects at once.
Object labeling in videos: our system uses a bounding box to support annotating single objects in a video frame such as cars, people, street signals, etc
Transcriptions for videos: supports manual transcription on videos frame by frame, in real time. This function is applied to new videos uploaded to the platform by default.
Action highlight for videos: usually applied for sport videos. Using machine learning and visual recognition, our action highlight function enrichs the viewers’ experience watching sport highlights.
Facial landmark in images: a well-trained labeling team to detect key facial structure and localizes regions of a face manually to ensure the highest accuracy.

Object labeling in images: uses 2D bounding box, polygon and segmentation tool to annotate a single or multiple objects in an image.
OCR on images: Annotate words on image

2. Labelling Tool Access

The labelling team can start a project with three easy steps

After signing in BlueEye Data platform, an user will see all the projects which were shared with them:

Step 1: Click on Projects which were shared with you

Step 2: Go to Dataset page and click on the Dataset menu

Step 3: Access labelling tool

Way 1: Click on Start Labeling button on Dataset page

If you are a Reviewer/ Project Manager/ Admin, you can see it as **Start Review** instead. In this case, the system will automatically open the first file on the first dataset on the Labeling tool.

Way 2: Click on Start Labeling button on a specific dataset page

You can see Start Labeling button after navigating to one of the datasets listed above.

If you are a Reviewer/ Project Manager/ Admin, you can see it as **Start Review** instead. In this case, the system will automatically open the first file on the first dataset on the Labeling tool.