Skip to content

Data Augmentation for Object Detection #7

Description

@romeokienzler

Feature

A data preprocessing step randomly imprinting object images from a pool to training data

Story / Description

OD often lacks sufficient training data as annotating labels (bboxes) is a tedious process. Instead of annotating real objects on the image one also can put images of real objects on images and annotate them automatically.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Labels

No labels
No labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions