Labeling Standards

Image Annotation Fields

Annotation Group	Field	Details
Image Identification Fields	image name	The filename of the image (e.g., image1.jpg).
Actual (Ground Truth) Annotations	category	The true category for the image.
Actual (Ground Truth) Bounding Box Annotations	x_center	X center coordinate for the bounding box.
Actual (Ground Truth) Bounding Box Annotations	y_center	Y center coordinate for the bounding box.
Actual (Ground Truth) Bounding Box Annotations	width	Width of the bounding box.
Actual (Ground Truth) Bounding Box Annotations	height	Height of the bounding box.
Predicted Annotations	predicted_category	The category label output by the model.
Predicted Annotations	predicted_probability_score	The confidence score of the prediction.
Predicted Bounding Box Annotations	predicted_x_center	X center coordinate for the predicted bounding box.
Predicted Bounding Box Annotations	predicted_y_center	Y center coordinate for the predicted bounding box.
Predicted Bounding Box Annotations	predicted_width	Width of the predicted bounding box.
Predicted Bounding Box Annotations	predicted_height	Height of the predicted bounding box.

On the labeling dataset, this will look like the following table and images, where the yellow box could be the actual (ground truth) labels, and the blue box could be the model predictions:

Image Name	Actual Category Label	Actual x center	Actual y center	Actual width	Actual height	Predicted Category Label	Predicted Probability Score	Predicted x center	Predicted y center	Predicted width	Predicted height
img10.jpg	ACV	2100	1450	3500	2500	ACV	0.97	2050	1450	3000	2800
img4.jpg	Osprey	575	400	1100	500	Osprey	0.96	540	370	850	500