Labelbox documentation

Core concepts

This section provides explanations of the core concepts and functionalities of Labelbox. Included in this section are diagrams, conceptual descriptions, methodology breakdowns, and other information for developing a more thorough understanding of how Labelbox works.

cloud_config_w_phoenix__2_.png

See Access & storage to learn how Labelbox accesses your data based on the deployment configuration you choose.

See Labelbox on-prem for details and requirements for using Labelbox on-premises.

dataset-gallery-2.png

See Dataset management to learn where to view and manage your datasets in Labelbox.

ontology-diagram.png

See Ontology framework to learn how ontologies are structured and how they are used.

model-assisted-labeling.png

See Model-assisted labeling to learn about importing annotations as pre-labels and why this workflow is useful.

image-editor.png

Explore the tools available in the Labelbox Editor for each asset type.

queue-system-2.png

See Queue system to learn about the reservation system and the parameters you can use to customize the label queue.

issues-and-comments.png

See Issues & comments to learn about why issues are useful, permissions associated with creating/modifying issues, and best practices.

open-review-label-browser-2.png

See the Review section to learn about open review vs queue-based review and things to be aware of when enabling the review step for your project.

account-usage.png

See Metrics to learn where to find metrics for annotation usage, labels created, and team performance.

See Member roles & permissions to see a breakdown of the permissions associated with organization roles, project-based roles, and workforce members.

See Benchmarks & Consensus methodology to learn how Labelbox calculates the Benchmarks & Consensus scores for each annotation type.