公开(公告)号：US20230186601A1

公开(公告)日：2023-06-15

申请号：US17912238

申请日：2021-03-17

Applicant: SeeChange Technologies Limited

Inventor： Timothy David HARTLEY ,  David Charles MCCAFFREY ,  Michael Andrew PALLISTER ,  Nimesh Naresh PATEL

IPC: G06V10/764 ,  G06K7/14

CPC classification number: G06V10/764 ,  G06K7/1447

Abstract: Apparatus having model-based machine learning and inferencing logic for controlling object transfer, comprises: an image input component to receive image data derived from a captured image of an object; a captured image classifier to 5 generate a first classification of the object by activating a trained model to analyse the image data; an input component to receive an object identifier for the object; an object identification classifier to generate a second classification of the object according to the object identifier; matching logic to detect failure to reconcile the first and second classification; heuristic logic responsive to the matching logic to 10 determine a causal factor in the failure; and training logic, operable when the heuristic logic determines that a causal factor in the failure to reconcile is a deficient first classification, to provide model training input comprising the image data and the object identifier to the model-based machine learning logic.

公开(公告)号：US20230177391A1

公开(公告)日：2023-06-08

申请号：US17911558

申请日：2021-03-17

Applicant: SEECHANGE TECHNOLOGIES LIMITED

Inventor： David PACKWOOD ,  Michael Andrew PALLISTER ,  Ariel Edgar RUIZ-GARCIA ,  Nimesh Naresh PATEL ,  Eleftherios FANIOUDAKIS

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: Provided is machine learning apparatus comprising: a dataset for input to a training procedure of a machine learning model; data capture logic operable to capture from an object at least one datum for inclusion in the dataset; association logic operable to derive an additional characteristic of the object; annotator logic operable in response to the data capture logic and the association logic to create an annotation linking the additional characteristic with the at least one datum; storage logic operable to store the or each datum with an associated annotation in the dataset; and input logic to supply the dataset as machine learning input.