In order to evaluate the accuracy of a given AI model inside the Benchmark application, your data must be labeled properly. Data labeling allows you to annotate each piece of content within your dataset so it can be recognized by models and used for predictions.

The Benchmark application makes this process of human-led labeling, quick and efficient by providing an intuitive and integrated solution.

Key Features

  • Labeling support for Video, Image, and Audio files

  • Feature-rich editor to increase labeling speed and accuracy

  • Custom classes and attributes to help organize and classify labels

  • Data specific annotation types to help label data properly

  • Time-based text editor

  • Keyframe video annotation

Label and Annotation Editing Tool

To access the Label and Annotation tool, navigate to the dataset you would like to annotate. To begin labeling your assets select from a single asset from the Assets tab or click the Label and Annotate option in your action bar at the top right of your page.

The Label and Annotation interface will automatically adjust to display the correct features for your Dataset content.

Did this answer your question?