Orientation (computer vision)

About: Orientation (computer vision) is a research topic. Over the lifetime, 17196 publications have been published within this topic receiving 358181 citations.

Method and system for visualizing an object

Abstract: In a method and a system for visualizing the position and orientation of an object that is penetrating, or that has penetrated, into a subject, a first set of image data are produced from the interior of the subject before the object has penetrated into the subject, a second set of image data are produced from the interior of the subject during or after the penetration of the object into the subject, the sets of image data are connected and are superimposed to form a fused set of image data, and an image obtained from the fused set of image data is displayed. The system has an x-ray computed tomography apparatus, and an x-ray apparatus, and/or an ultrasound apparatus for producing the first and second sets of data, respectively.

Optimal estimator model for human spatial orientation

Abstract: A model is being developed to predict pilot dynamic spatial orientation in response to multisensory stimuli Motion stimuli are first processed by dynamic models of the visual, vestibular, tactile, and proprioceptive sensors Central nervous system function is then modeled as a steady-state Kalman filter which blends information from the various sensors to form an estimate of spatial orientation Where necessary, this linear central estimator has been augmented with nonlinear elements to reflect more accurately some highly nonlinear human response characteristics Computer implementation of the model has shown agreement with several important qualitative characteristics of human spatial orientation, and it is felt that with further modification and additional experimental data the model can be improved and extended Possible means are described for extending the model to better represent the active pilot with varying skill and work load levels

Apparatus and method for generating image representations of a body utilizing an ultrasonic imaging subsystem and a three-dimensional digitizer subsystem

Abstract: The disclosure is directed to an apparatus and method for producing a three-dimensional image representation of a body. The body can be animate or inanimate; i.e. the invention can be used to obtain 3D image representations of for example, parts of the human body or any object that one wishes to view or measure. The 3D image representations can be used to produce 2D displays of sections, contours, etc., or displays having 3D perspective, such as wire-frame type illustrations. This facilitates automatic computation of areas or volumes. In a disclosed embodiment, an ultrasonic imaging subsystem is provided for producing signals representative of two-dimensional images of sections of the body, the subsystem including a scanning transducer that is moveable to determine the section of the body to be imaged. The image representative signals are stored as arrays of digital pixel values. A three-dimensional acoustic digitizer subsystem is provided for deriving and storing information representative of the position and orientation of the transducer during the scanning of an associated section of the body. A voxel space storage is provided for storing a three-dimensional array of voxel values. The arrays of digital pixel values are projected into the voxel space storage, the voxel locations which correspond to projected pixel locations of a given pixel array being determined as a function of the stored position and orientation information associated with the section of the body from which the given pixel array was obtained.

Visual Camera Re-Localization from RGB and RGB-D Images Using DSAC

Abstract: We describe a learning-based system that estimates the camera position and orientation from a single input image relative to a known environment. The system is flexible w.r.t. the amount of information available at test and at training time, catering to different applications. Input images can be RGB-D or RGB, and a 3D model of the environment can be utilized for training but is not necessary. In the minimal case, our system requires only RGB images and ground truth poses at training time, and it requires only a single RGB image at test time. The framework consists of a deep neural network and fully differentiable pose optimization. The neural network predicts so called scene coordinates, i.e. dense correspondences between the input image and 3D scene space of the environment. The pose optimization implements robust fitting of pose parameters using differentiable RANSAC (DSAC) to facilitate end-to-end training. The system, an extension of DSAC++ and referred to as DSAC*, achieves state-of-the-art accuracy an various public datasets for RGB-based re-localization, and competitive accuracy for RGB-D-based re-localization.