Use Roboflow Annotate
An overview of the Labeling Interface for Roboflow's Annotation Tool, including shortcut keys.
Last updated
An overview of the Labeling Interface for Roboflow's Annotation Tool, including shortcut keys.
Last updated
You can access the labeling interface by selecting an image from the Assign or Dataset pages on the Roboflow dashboard.
On the right-hand side of the labeling interface, you will find the toolbar. The toolbar has many features you can use for annotating images.
In this document, we talk through how to use the following features:
Drag and select
Bounding box annotation tool
Polgyon annotation tool
Smart polygon
Label assist
Zoom tool
Represented by a hand icon, this feature allows you to select, edit, and drag individual annotations.
Single-click an existing bounding box to select it. Once selected, you can change a bounding box's size with the circular white handles that appear on its corners and on each side. Or use the class editor to change the box's label.
Drag a box to move it.
Drag the background to pan.
Click the background to deselect all boxes.
The bounding box annotation tool (represented by a rectangular box icon) allows you to draw new bounding-box annotations. In this mode, you will see crosshairs that will help you determine where to start drawing.
Click and drag across an image to create a new annotation, then use the Class Selector to choose its label.
The polygon annotation tool allows you to draw new polygonal annotations. In this mode, you will see crosshairs that will help you determine where to start drawing.
Click on the image around objects of interest to create an enclosed polygon annotation, then use the Class Selector to choose its label.
Smart Polygon allows you to draw new Smart Polygon annotations.
In this mode, you will see a green dot when you are selecting a new area of interest (new label); a red dot when selecting areas to remove from the area of interest (parts of the object or image that you don't want to label/enclose with the polygon); and options to adjust the polygon by Convex Hull, Smooth, and Complex settings.
Smart Polygon is particularly useful for (Instance and Semantic) Segmentation projects, however, you may see performance boosts in Object Detection models when labeling with Smart Polygons on Roboflow.
Label Assist allows for the use of a Public Model (such as the COCO model) or one of your own dataset versions that was trained with Roboflow Train, for the automatic application of bounding box labels to images in your Annotation queue.
Mark null (Null annotation) is to be used for the "labeling" of background, or null, images. This setting can also be used to clear all annotations from an image, or to mark the image as Unannotated. To learn more about null annotations, check out our guide "The Difference Between Missing and Null Annotations".
While in Bounding Box (B), Polygon (P), or Smart Polygon (S) mode:
Undo reverts the previous action.
Redo reverses a previously undone action.
Repeat Previous reapplies label(s) on an image in the same location(s) as the last annotated image
When an image is selected, the Class Selector will appear. It contains the following options for choosing the label of a bounding box:
Textfield to create a new class or filter existing classes.
Buttons to save or discard your changes.
Class List of the existing classes in the dataset (filtered by the text field and with the active option highlighted in purple) and, sometimes, a "Create class" option if the text you typed does not match an existing class.
The zoom tool found at the bottom left of the screen.
Zoom in and out to fit more of the image on your screen at one time or to get a closer look for more detailed editing. There is also an option to "lock" the zoom to a specified percentage, or reset the zoom to fit the entirety of the image within the Annotation Tool's viewport.
Note that if you select the "Zoom Lock" option, all images will appear at this zoom-level. Deselect, or unlock, the lock to remove Zoom Lock.
Annotations (abbreviated Annots in the dashbaord) show which classes are present and not present in an image, what color their boxes are, and layering of labels. The Annotations drawer includes Tags which can be used to help with organizing, filtering or sorting through images in datasets.
Attributes represent information about an image including its dimensions, last-modified time, and whether it is in this dataset's training, validation, or test set.
