Data annotation for every industry.
Same rigour, very different data.
A shelf photo and a drone pass over a transmission line need the same careful process and completely different judgement. These pages show what we label in each domain, which annotation type it starts from, and what usually goes wrong.
- Industries
- 12
- Annotation types
- 7
- Quote
- Human-reviewed
- Delivery
- 5 formats
Browse
Pick the one closest to your data
Manufacturing
Instance segmentationDefect detection and assembly verification on the line.
scratch · dent · weld_gap
Retail & E-commerce
Object detectionShelf audits, planogram checks and product recognition.
product_facing · price_tag · empty_slot
Agriculture
Instance segmentationCrop, weed and yield analysis from drone or ground imagery.
crop · weed · diseased_leaf
Healthcare & Medical Imaging
Instance segmentationRegion-of-interest and cell-level annotation for clinical research.
lesion · nodule · cell
Logistics & Warehousing
Object detectionPallet, parcel and dock-door tracking across a facility.
pallet · parcel · forklift
Construction & Site Safety
Object detectionPPE compliance, progress tracking and hazard spotting.
person · helmet · hi_vis_vest
Transportation & Traffic
Object detectionVehicle counting, ANPR support and junction analytics.
car · truck · bus
Energy & Utilities
Object detectionAsset inspection on lines, panels and pipelines.
insulator · corrosion · cracked_panel
Security & Surveillance
Object detectionIntrusion, loitering and object-left-behind detection.
person · vehicle · bag
Automotive
Object detectionIn-cabin monitoring, ADAS datasets and paint inspection.
vehicle · lane_marking · traffic_sign
Sports & Media
Keypoints & posePlayer tracking, pose analysis and broadcast tagging.
player · ball · referee
Robotics & Drones
Instance segmentationGrasp targets, navigation cues and aerial survey.
graspable_object · obstacle · landing_pad
By task
Or start from the annotation type
Bounding boxes
Tight 2D boxes around every object, class by class.
Polygons & instance segmentation
Pixel-accurate outlines and masks per object.
Semantic segmentation
Every pixel in the image assigned a class.
Keypoints & pose
Ordered joints and landmarks on people and objects.
Oriented boxes
Rotated boxes for aerial, shelf and document imagery.
Image classification
Image-level tags and attributes, at volume.
Video & object tracking
Frame-by-frame tracks with persistent IDs.
Not sure which one you are?
Most projects fit more than one of these. Send us a sample and we will tell you which annotation types fit your data and what it costs to label.