公开(公告)号：US11734561B2

公开(公告)日：2023-08-22

申请号：US17571737

申请日：2022-01-10

Applicant: Intel Corporation

Inventor： Anthony Rhodes ,  Manan Goel

IPC: G06N3/08 ,  G06V20/40 ,  G06V10/764 ,  G06V10/82 ,  G06V10/44 ,  G06V40/10 ,  G06V10/62

CPC classification number: G06N3/08 ,  G06V10/454 ,  G06V10/764 ,  G06V10/82 ,  G06V20/40 ,  G06V20/41 ,  G06V20/46 ,  G06V40/103 ,  G06V10/62

Abstract: An apparatus, method, system and computer readable medium for video tracking. An exemplar crop is selected to be tracked in an initial frame of a video. Bayesian optimization is applied with each subsequent frame of the video by building a surrogate model of an objective function using Gaussian Process Regression (GPR) based on similarity scores of candidate crops collected from a search space in a current frame of the video. A next candidate crop in the search space is determined using an acquisition function. The next candidate crop is compared to the exemplar crop using a Siamese neural network. Comparisons of new candidate crops to the exemplar crop are made using the Siamese neural network until the exemplar crop has been found in the current frame. The new candidate crops are selected based on an updated surrogate model.

公开(公告)号：US20220397962A1

公开(公告)日：2022-12-15

申请号：US17825872

申请日：2022-05-26

Applicant: Intel Corporation

Inventor： Manan Goel ,  Saurin Shah ,  Lakshman Krishnamurthy ,  Steven Xing ,  Matthew Pinner ,  Kevin James Doucette

IPC: G06F3/01 ,  G06F3/03 ,  G10H1/00 ,  G06F3/16 ,  G06F3/042 ,  G06V40/20

Abstract: Gesture-controlled virtual reality systems and methods of controlling the same are disclosed herein. An example apparatus includes an on-body sensor to output first signals associated with at least one of movement of a body part of a user or a position of the body part relative to a virtual object and an off-body sensor to output second signals associated with at least one of the movement or the position relative to the virtual object. The apparatus also includes at least one processor to generate gesture data based on at least one of the first or second signals, generate position data based on at least one of the first or second signals, determine an intended action of the user relative to the virtual object based on the position data and the gesture data, and generate an output of the virtual object in response to the intended action.

公开(公告)号：US20220130130A1

公开(公告)日：2022-04-28

申请号：US17571737

申请日：2022-01-10

Applicant: Intel Corporation

Inventor： Anthony Rhodes ,  Manan Goel

IPC: G06V10/46 ,  G06N3/08 ,  G06V10/426 ,  G06V20/40

Abstract: An apparatus, method, system and computer readable medium for video tracking. An exemplar crop is selected to be tracked in an initial frame of a video. Bayesian optimization is applied with each subsequent frame of the video by building a surrogate model of an objective function using Gaussian Process Regression (GPR) based on similarity scores of candidate crops collected from a search space in a current frame of the video. A next candidate crop in the search space is determined using an acquisition function. The next candidate crop is compared to the exemplar crop using a Siamese neural network. Comparisons of new candidate crops to the exemplar crop are made using the Siamese neural network until the exemplar crop has been found in the current frame. The new candidate crops are selected based on an updated surrogate model.

公开(公告)号：US20200026954A1

公开(公告)日：2020-01-23

申请号：US16586671

申请日：2019-09-27

Applicant: Intel Corporation

Inventor： Anthony Rhodes ,  Manan Goel

IPC: G06K9/46 ,  G06K9/00 ,  G06N3/08

Abstract: An apparatus, method, system and computer readable medium for video tracking. An exemplar crop is selected to be tracked in an initial frame of a video. Bayesian optimization is applied with each subsequent frame of the video by building a surrogate model of an objective function using Gaussian Process Regression (GPR) based on similarity scores of candidate crops collected from a search space in a current frame of the video. A next candidate crop in the search space is determined using an acquisition function. The next candidate crop is compared to the exemplar crop using a Siamese neural network. Comparisons of new candidate crops to the exemplar crop are made using the Siamese neural network until the exemplar crop has been found in the current frame. The new candidate crops are selected based on an updated surrogate model.

公开(公告)号：US20200026928A1

公开(公告)日：2020-01-23

申请号：US16584709

申请日：2019-09-26

Applicant: Intel Corporation

Inventor： Anthony Rhodes ,  Manan Goel

IPC: G06K9/00 ,  G06T7/20 ,  G06T7/174 ,  G06T7/00 ,  G06T7/70 ,  G06K9/62

Abstract: Techniques related to automatically segmenting video frames into per pixel dense object of interest and background regions are discussed. Such techniques include applying a segmentation convolutional neural network (CNN) to a CNN input including a current video frame, a previous video frame, an object of interest indicator frame, a motion frame, and multiple feature frames each including features compressed from feature layers of an object classification convolutional neural network as applied to the current video frame to generate candidate segmentations and selecting one of the candidate segmentations as a final segmentation of the current video frame.

公开(公告)号：US20180315405A1

公开(公告)日：2018-11-01

申请号：US15581957

申请日：2017-04-28

Applicant: Intel Corporation

Inventor： Saurin Shah ,  Swarnendu Kar ,  Lakshman Krishnamurthy ,  Arthur D. Webb ,  Manan Goel

IPC: G10H1/00 ,  G06F3/01 ,  G06F3/16 ,  G06T11/60

Abstract: Systems and methods may be used to provide effects corresponding to movement of instrument objects or other objects. A method may include receiving sensor data from an object based on movement of the object, recognizing a gesture from the sensor data, and determining an effect, such as a visualization or audio effect corresponding to the gesture. The method may include causing the effect to be output in response to the determination.

公开(公告)号：US10049650B2

公开(公告)日：2018-08-14

申请号：US15275252

申请日：2016-09-23

Applicant: INTEL CORPORATION

Inventor： Manan Goel ,  Saurin Shah ,  Lakshman Krishnamurthy ,  Francis M. Tharappel ,  Swarnendu Kar ,  Suresh V. Golwalkar

IPC: G10H1/00 ,  G06K7/10

Abstract: The present disclosure describes a number of embodiments related to devices, systems, and methods locating a an object using ultra-wide band (UWB) radio transceivers embedded in carpet or other flexible material that may be rolled up and moved to various locations. Once in a location, the carpet may be unrolled and the multiple embedded radio transceivers may receive a signal from a tag attached to the object sending UWB radio signals. Based on the signals received by the UWB radio transceivers, various processes including time-difference on arrival, time-of-flight, and phase shift may be used to determine the location or the movement of the object.

公开(公告)号：US09939910B2

公开(公告)日：2018-04-10

申请号：US14978822

申请日：2015-12-22

Applicant: Intel Corporation

Inventor： Saurin Shah ,  Narayan Sundararajan ,  Manan Goel ,  Brian K. Vogel ,  Jason Blanchard ,  Jason Wright ,  Lakshman Krishnamurthy ,  Swarnendu Kar

IPC: G06F3/01 ,  G10H1/00 ,  G10H1/34

CPC classification number: G06F3/017 ,  G06F3/011 ,  G10H1/0008 ,  G10H1/0066 ,  G10H1/348 ,  G10H2220/201 ,  G10H2220/321 ,  G10H2220/395 ,  G10H2240/211

Abstract: Processing techniques and device configurations for performing and controlling output effects at a plurality of wearable devices are generally described herein. In an example, a processing technique may include receiving, at a computing device, an indication of a triggering gesture that occurs at a first wearable device, determining an output effect corresponding to the indication of the triggering gesture, and in response to determining the output effect, transmitting commands to computing devices that are respectively associated with a plurality of wearable devices, the commands causing the plurality of wearable devices to generate the output effect at the plurality of wearable devices. In further examples, output effects such as haptic feedback, light output, or sound output, may be performed by the plurality of wearable devices, associated computing devices, or other controllable equipment.

公开(公告)号：US12045986B2

公开(公告)日：2024-07-23

申请号：US17131525

申请日：2020-12-22

Applicant: Intel Corporation

Inventor： Anthony Rhodes ,  Manan Goel ,  Ke Ding

IPC: G06T7/00 ,  G06F18/20 ,  G06N3/045 ,  G06T7/10 ,  G06T7/11 ,  G06V10/44 ,  G06V10/764 ,  G06V10/82

CPC classification number: G06T7/11 ,  G06F18/285 ,  G06N3/045 ,  G06T7/10 ,  G06V10/454 ,  G06V10/764 ,  G06V10/82

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed for interactive image segmentation. An example apparatus includes an inception controller to execute an inception sublayer of a convolutional neural network (CNN) including two or more inception-atrous-collation (IAC) layers, the inception sublayer including two or more convolutions including respective kernels of varying sizes to generate multi-scale inception features, the inception sublayer to receive one or more context features indicative of user input; an atrous controller to execute an atrous sublayer of the CNN, the atrous sublayer including two or more atrous convolutions including respective kernels of varying sizes to generate multi-scale atrous features; and a collation controller to execute a collation sublayer of the CNN to collate the multi-scale inception features, the multi-scale atrous features, and eidetic memory features.

公开(公告)号：US11928753B2

公开(公告)日：2024-03-12

申请号：US16773715

申请日：2020-01-27

Applicant: INTEL CORPORATION

Inventor： Anthony Rhodes ,  Manan Goel

IPC: G06T1/20 ,  G06F18/241 ,  G06T3/40 ,  G06T7/11 ,  G06T7/174 ,  G06T9/00 ,  G06V10/26 ,  G06V10/764 ,  G06V20/40

CPC classification number: G06T1/20 ,  G06F18/241 ,  G06T3/4046 ,  G06T7/11 ,  G06T7/174 ,  G06T9/002 ,  G06V10/26 ,  G06V10/764 ,  G06V20/46 ,  G06V20/49 ,  G06T2207/10016 ,  G06T2207/20221

Abstract: Techniques related to automatically segmenting video frames into per pixel fidelity object of interest and background regions are discussed. Such techniques include applying tessellation to a video frame to generate feature frames corresponding to the video frame and applying a segmentation network implementing context aware skip connections to an input volume including the feature frames and a context feature volume corresponding to the video frame to generate a segmentation for the video frame.