2D Bounding Boxes services · 01 / 10
2D Bounding Boxes
Tight, consistent boxes for object detection, including occluded and truncated objects.
Detection models are only as sharp as the edges you train them on. We draw pixel-tight boxes against a written spec that settles the hard cases up front: occlusion, truncation, reflections, crowds and minimum object size.
Typical use cases
- Retail shelf & SKU detection
- Traffic & ADAS perception
- Drone and aerial inspection
- Warehouse and logistics vision
- Security and access monitoring
Sample deliverable
What the output looks like.
Synthetic demonstration scene. No client data.
Variants
Axis-aligned boxes
Standard detection boxes with class, occlusion and truncation attributes.
Rotated boxes
Oriented boxes for aerial imagery, documents and packed objects.
Crowd & group flags
Explicit handling for dense scenes so your loss function sees the truth.
QA metric
Mean IoU vs gold set
≥ 0.92
Measured on a blind gold sample in every batch.
Export formats
- COCO JSON
- YOLO TXT
- Pascal VOC
- CVAT XML
- Custom schema
How every project runs
- Scope call & sample review
- Written guideline & gold set
- Fixed-fee pilot with QA report
- Production in weekly batches
Next step
Scope your project with
an annotation lead.
A 30-minute call with the person who would run your project. We review a sample of your data, recommend an approach and send a fixed-price proposal within two business days.