A robotic system includes a robot having an associated workspace; a vision sensor constructed to obtain a 3D image of a robot scene including a workpiece located in the workspace; and a control system communicatively coupled to the vision sensor and to the robot. The control system is configured to execute program instructions to filter the image by segmenting the image into a first image portion containing substantially only a region of interest within the robot scene, and a second image portion containing the balance of the robot scene outside the region of interest; and by storing image data associated with the first image portion. The control system is operative to control movement of the robot to perform work, on the workpiece based on the image data associated with the first image portion.

A method for performing segmentation on CT image data of a spine is provided. The method includes retrieving the CT image data of the spine, detecting an estimated position of at least four pedicle regions associated with a plurality of vertebras, determining a pose for each of at least two vertebrae. The method further includes performing a first segmentation process on the CT image data with a shape model to generate a first segmentation and performing a second segmentation process on the CT image data using a first neural network to generate a second segmentation of the at least two vertebrae. The method also includes mapping the shape model to the second segmentation, applying landmarks from the shape model to the second segmentation using the mapping, and overlaying a segmentation mask based on the second segmentation over the CT image data.

A method and device for synthesizing a first face in a first image, by determining a first occluded part of the first face that is occluded by an occluding object; determining a first visible part of the first face from the first occluded part; calculating first attributes representative of the first visible part; obtaining first parameters representative of an appearance of the first face by applying a regressor to the first attributes, the regressor modelling a correlation between second attributes representative of second visible parts of a plurality of second faces in second images and second parameters representative of an appearance model of the plurality of second faces; and synthesizing the first face using the first parameters.

A method of tracking faces in an image stream with a digital image acquisition device includes receiving images from an image stream including faces, calculating corresponding integral images, and applying different subsets of face detection rectangles to the integral images to provide sets of candidate regions. The different subsets include candidate face regions of different sizes and/or locations within the images. The different candidate face regions from different images of the image stream are each tracked.

Systems and methods for anatomical structure segmentation in medical images using multiple anatomical structures, instructions and segmentation models.

Intelligent image parsing for anatomical landmarks and/or organs detection and/or segmentation is provided. A state space of an artificial agent is specified for discrete portions of a test image. A set of actions is determined, each specifying a possible change in a parametric space with respect to the test image. A reward system is established based on applying each action of the set of actions and based on at least one target state. The artificial agent learns an optimal action-value function approximator specifying the behavior of the artificial agent to maximize a cumulative future reward value of the reward system. The behavior of the artificial agent is a sequence of actions moving the agent towards at least one target state. The learned artificial agent is applied on a test image to automatically parse image content.

Intelligent image parsing for anatomical landmarks and/or organs detection and/or segmentation is provided. A state space of an artificial agent is specified for discrete portions of a test image. A set of actions is determined, each specifying a possible change in a parametric space with respect to the test image. A reward system is established based on applying each action of the set of actions and based on at least one target state. The artificial agent learns an optimal action-value function approximator specifying the behavior of the artificial agent to maximize a cumulative future reward value of the reward system. The behavior of the artificial agent is a sequence of actions moving the agent towards at least one target state. The learned artificial agent is applied on a test image to automatically parse image content.

Supported by artificial intelligence, an object is classified in an X-ray projection image. A 3D representation as well as a localization of the classified object can be determined by matching a model of the classified object to a visualization of the classified object in the X-ray image.