Capability03.05Systems
Computer Vision
Systems that see what’s in front of them and report what matters.
Computer vision earns its keep on the line, in the yard and around machines: finding the defect, following the parcel, noticing someone inside a safety zone. We build vision systems to report what matters, to say when they are unsure, and to keep faces and identities out of anything that doesn’t need them.
Drag a part off the line: the tracker keeps its ID. Arrow keys cycle the detections.
Live view is off on this device.
Includes
- Detection, segmentation and multi-object tracking
- Visual quality inspection
- Retail and site analytics, privacy by design
- Safety-zone monitoring without identifying people
- Pose estimation and robotics vision
- Synthetic training data
Applications
- Surface-defect inspection on a production line, with uncertain cases sent to a person
- Tracking parcels and containers through a yard or warehouse
- Monitoring safety zones around machines without storing faces
Under the hood
- Object detection and instance segmentation
- Multi-object tracking with persistent IDs
- Few-shot anomaly detection for defects
- Pose estimation
- Edge inference on embedded GPUs
- Synthetic data and domain randomisation
How we test it
We measure precision and recall on held-out footage from your own site, broken down by lighting, angle, product variant and shift. Thresholds are set with the people who handle the exceptions, and every uncertain case goes to a person, then back into the training set.
Have something like this in mind?
Bring the problem in its messiest form. Measuring it is our job. We’ll tell you honestly whether AI is the right way to solve it.