Use Ground Truth for Training AI Models
Labeled ground truth data is essential for training supervised AI models across a wide range of computer vision tasks, including object detection, semantic segmentation, image classification, and video activity recognition. Computer Vision Toolbox™ provides tools to help you prepare labeled ground truth data for deep learning training by selecting relevant labels, modifying file paths, merging ground truth objects, and organizing data sets for training and evaluation.
The Image Labeler and Video Labeler apps export labeled
ground truth data in the form of a groundTruth object. To generate
training data sets by converting labeled ground truth into formats compatible
with AI models, use functions like
objectDetectorTrainingData,
pixelLabelTrainingData, and
sceneLabelTrainingData. These functions support object
detection, segmentation, and classification tasks. For more information, see
Training Data for Object Detection and Semantic Segmentation and Postprocess Exported Labels for Instance Segmentation Training. You can
also create blocked image representations using
polyToBlockedImage enabling you to efficiently process
large-scale images.
To select specific labels from ground truth data, and filter and organize
annotations based on task requirements, use functions like
selectLabelsByGroup,
selectLabelsByType, or
selectLabelsByName. The toolbox also supports
post-processing of labeled data using functions such as
merge to combine multiple ground truth objects,
changeFilePaths to update data set references, and
gatherLabelData to extract label information. For video
data, utilities like writeVideoScenes and
sceneTimeRanges help manage scene-level
annotations.
To share and review image labels with colleagues, consider creating a team project within the Image Labeler app. For more details, see Get Started with Team-Based Labeling.
Apps
| Image Labeler | Label images for computer vision applications |
| Video Labeler | Label video for computer vision applications |
Functions
Topics
- Elements of Ground Truth Objects
Understand how to save and pass data using a ground truth data object.
- Share and Store Labeled Ground Truth Data
Share and store labeled ground truth data exported from labeling apps.
- How Labeler Apps Store Exported Pixel Labels
Learn how the labeling apps store pixel label data.
- Training Data for Object Detection and Semantic Segmentation
Create training data for object detection or semantic segmentation using the Image Labeler or Video Labeler.
- Postprocess Exported Labels for Instance Segmentation Training
Postprocess exported ground truth labels and create training datastore for training instance segmentation networks such as SOLOv2 or Mask R-CNN.
- Datastores for Deep Learning (Deep Learning Toolbox)
Learn how to use datastores in deep learning applications.
- Get Started with Image Preprocessing and Augmentation for Deep Learning
Preprocess data for deep learning applications with deterministic operations such as resizing, or augment training data with randomized operations such as random cropping.
Featured Examples
Create Instance Segmentation Training Data From Ground Truth
Create instance segmentation training data from a groundTruth object. To train a Mask R-CNN and SOLO v2 network using the trainMaskRCNN and trainSOLOv2 functions, respectively, format your input data as a 1-by-4 cell array containing the RGB training image, bounding boxes, instance labels, and instance masks. Convert your polygon ground data into a set of binary instance masks and axis-aligned rectangles for training.
