Abstract: A method of performing quality assessment in a process of manufacture of or processing of a product is provided. The method comprises: providing a specification (500) for the product, generating, from the specification, synthetic data (400) representative of the appearance of the product when conforming to the specification and, separately, the appearance of the product when defective. An Al model can be trained (420) using the synthetic data to distinguish between acceptable products and defective products, for use of the trained Al model on images of real products in a manufacturing or processing facility (450). On-line and off-line inspection systems are described. A sensor (115) captures raw data about an object and this, in conjunction with the trained Al model allows the system to distinguish between acceptable objects and defective objects, and to perform other tasks such as measuring.

Abstract: Systems and methods for determining pose using a trained neural network are described, whereby a user device receives image data of a 3-dimensional (“3D”) marker affixed to a 3D object to be tracked, provides a set of input data derived from the image data to a neural network stored on the user device, and generates a pose descriptor indicative of estimated pose of the 3D marker based on output of the neural network produced in response to receiving the set of input data. The 3D marker comprises a first surface to convey radiation in a first direction, and a second surface to convey radiation in a second direction different to the first direction, whereby the image processing system determines object pose from captured image data of at least a portion of the radiation conveyed from the first and/or second surface of the 3D marker affixed to the 3D object.

Abstract: Systems and methods for determining pose using a trained neural network are described, whereby a user device receives image data of a marker affixed to an object to be tracked, provides a set of input data derived from the image data to a neural network stored on the user device, and generates a pose descriptor indicative of estimated pose of the marker based on output of the neural network produced in response to receiving the set of input data. The marker comprises a first surface to convey radiation in a first direction, and a second surface to convey radiation in a second direction different to the first direction, whereby the image processing system determines object pose from captured image data of at least a portion of the radiation conveyed from the first and/or second surface of the marker affixed to the object. Other embodiments are also described and claimed.