Intel Corporation

University of Pennsylvania

Rutgers, The State University of New Jersey

Invensense, Inc.

Jianbo Shi

Daniel C Middleton

Anant Madabhushi

Steven Saeed Nasiri

David Sachs

Kabeer R Manchanda

Varun Nasery

Gilson Goncalves De Lima

Michael D Feldman

Todd A Keaffaber

Mark Rosen

John Tomaszewski

Martin Meloun

Peter John Madsen

Vivek Paresh Buch

Philip Kim

Christopher Bernard

Yedong Niu

Kishore Moturi

A method for generating and providing artificial intelligence assisted surgical guidance includes analyzing video images from surgical procedure and training a neural network to identify at least one of anatomical objects, surgical objects, and tissue manipulation in video images. The method includes receiving, by the neural network, a live feed of video images from the surgery. The method further includes classifying, by the neural network, at least one of anatomical objects, surgical objects, and tissue manipulation in the live feed of video images. The method further includes outputting, in real time, surgical guidance based on algorithms created using the classified at least on of anatomical objects, surgical objects, and tissue manipulations in the live feed of video images.

Methods, systems, and computer readable media for generating and providing artificial intelligence assisted surgical guidance

Techniques are provided for synthesis of transformed image views, based on a reference image, using depth information. The transformed image views may simulate a change in position or focal length of a camera that produced the reference image. An example system includes an image transformation circuit configured to transform the reference image corresponding to a first viewpoint, to a transformed image corresponding to a second viewpoint. The system also includes an inverse warping circuit configured to calculate a mapping from the pixels of the transformed image to corresponding pixels of the reference image. The system further includes a hole detection circuit configured to detect holes in the transformed image based on depth discontinuities between the reference and transformed images; and a hole filling circuit configured to in-fill the detected holes using a sampling of selected neighboring pixels from the reference image, to synthesize a view based on the transformed image.

Synthesis of transformed image views

Techniques for calibration of image capturing devices are described herein. The techniques include capturing one or more images at a first target, capturing one or more images at a second target, capturing one or more images at a third target, capturing one or more images at a fourth target, and calibrating a multi-camera system based on the captured images.

Multi-target camera calibration

Devices, systems and methods employing modular camera arrays are described. A two-dimensional array of cameras may be arranged in a non-rectangular array. A first camera, a second camera, and a third camera in the array may be located approximately equidistant from each other, and/or may be arranged approximately in an equilateral triangle, an isosceles triangle, a scalene triangle, and/or a right triangle.

Modular camera array

Various embodiments are generally directed to repeated use of wavelet filters with different directional orientations each time to detect distances to edges in different directions in deriving a disparity map from multiple image bitmaps. A method includes subjecting an image bitmap to a first-order wavelet filter a selected number of times, an orientation of the first-order wavelet filter altered each time to create a first set of intermediate maps; and subjecting the first image bitmap to a second-order wavelet filter the selected number of times, an orientation of the second-order wavelet filter altered each time to create a second set of intermediate maps; the first-order and second-order wavelet filters, as well as their orientations, selected to be mathematically related such that the first-order wavelet filter approximates a sine function and the second-order wavelet filter approximates a related cosine function, both masked by a Gaussian. Other embodiments are described and claimed.

Techniques for wavelet-based image disparity estimation

This invention relates to supervised or unsupervised classification of biological datasets. Specifically, the invention relates to the use of Graph Embedding as a method of reducing dimensionality thereby improving supervised classification of classes, both conventional and new ones.

Systems and methods for classification of biological datasets

Methods and apparatus for electronically stabilizing an image captured by a device including a motion detection unit are provided. In one implementation, the method includes capturing a first exposure of the image, and capturing a second exposure of the image including using the motion detection unit to ensure that the second exposure of the image has a pre-determined blur property. The second exposure is longer than the first exposure. The method further includes combining the second exposure of the image having the pre-determined blur property and the first exposure of the image to electronically stabilize the image captured by the device.

Method and apparatus for producing a sharp image from a handheld device containing a gyroscope

Growing community of inventors

Philadelphia, PA, United States of America

Jianbo Shi