Detection of hand gestures using gesture language discrete values

    公开(公告)号:US10599919B2

    公开(公告)日:2020-03-24

    申请号:US14985680

    申请日:2015-12-31

    摘要: Computer implemented method for detecting a hand gesture of a user, comprising: (a) Receiving sequential logic models each representing a hand gesture. The sequential logic model maps pre-defined hand poses and motions each represented by a hand features record defined by discrete hand values each indicating a state of respective hand feature. (b) Receiving a runtime sequence of runtime hand datasets each defined by discrete hand values scores indicating current state hand features of a user's moving hand which are inferred by analyzing timed images depicting the moving hand. (c) Submitting the runtime hand datasets and the pre-defined hand features records in SSVM functions to generate estimation terms for the runtime hand datasets with respect to the hand features records. (d) Estimating which of the hand gestures best matches the runtime sequence depicted in the timed images by optimizing score functions using the estimation terms for the runtime hand datasets.

    Structure and training for image classification

    公开(公告)号:US10460201B2

    公开(公告)日:2019-10-29

    申请号:US14985803

    申请日:2015-12-31

    摘要: A computer implemented method of training an image classifier, comprising: receiving training images data labeled according to image classes; selecting reference points of the images; and constructing a set of voting convolutional tables and binary features on a patch surrounding each reference point by performing, for each calculation stage: creating a voting table by: creating first candidate binary features; calculating a global loss reduction for each first candidate binary feature; selecting one first candidate binary feature having minimal global loss reduction; and repeating to select stage-size binary features; and performing a tree split using the voting table by: creating second candidate binary features; calculating a combined loss reduction for each stage-split size group of the second candidate binary features; selecting one of the groups having a maximal combined loss reduction; and creating a child-directing table using the selected binary features.

    TRANSFORM LIGHTWEIGHT SKELETON AND USING INVERSE KINEMATICS TO PRODUCE ARTICULATE SKELETON

    公开(公告)号:US20170193289A1

    公开(公告)日:2017-07-06

    申请号:US14985777

    申请日:2015-12-31

    IPC分类号: G06K9/00 G06T7/00

    摘要: A system of inverse reconstruction of a skeleton model of a hand, comprising: an imager adapted to capture at least one image of a hand; a memory storing a plurality of hand pose features records, each defined by a unique set of discrete pose values; a code store storing a code; at least one processor coupled to the imager, memory and program store for executing the stored code, the code comprising: code instructions to identify a group of discrete pose values from an analysis of the at least one image; code instructions to select a hand pose features record from the hand pose features records according to the group of discrete pose values; and code instructions to reconstruct a skeleton model of the hand in the hand pose from the hand pose features record based on a hand model which maps kinematic characteristics of a plurality of hand organs.

    Depth map correction using lookup tables

    公开(公告)号:US10230934B2

    公开(公告)日:2019-03-12

    申请号:US13917974

    申请日:2013-06-14

    发明人: Eyal Krupka

    摘要: Depth map correction using lookup tables is described. In an example depth maps may be generated that measure a depth to an object using differences in phase between light transmitted from a camera which illuminates the object and light received at the camera which has been reflected from the object. In various embodiments depth maps may be subject to errors caused by received light undergoing multiple reflections before being received by the camera. In an example a correction for an estimated depth of an object may be computed and stored in a lookup table which maps the amplitude and phase of the received light to a depth correction. In an example the amplitudes and frequencies of each modulation frequency may be to access lookup table which stores corrections for the depth of an object and which allows an accurate depth map to be obtained.

    Recognition of hand poses by classification using discrete values

    公开(公告)号:US09734435B2

    公开(公告)日:2017-08-15

    申请号:US14985741

    申请日:2015-12-31

    IPC分类号: G06K9/00 G06K9/62 G06K9/66

    摘要: Computer implemented method for computing a feature dataset classifying a pose of a human hand, comprising: (a) Selecting a global orientation category (GOC) defining a spatial orientation of a human hand in a 3D space by applying GOC classifying functions on a received image segment depicting the hand. (b) Identifying in-plane rotation by applying in-plane rotation classifying functions on the image segment, the in-plane rotation classifying functions are selected according to said GOC. (c) Aligning the image segment in a 2D plane according to the in-plane rotation. (d) Applying hand pose features classifying functions on the aligned image segment. Each one of the feature classifying functions outputs a current discrete pose value of an associated hand feature. (e) Outputting a features dataset defining a current discrete pose value for each of the hand pose features for classifying current hand pose of the hand.

    SUPPLEMENTING BIOMETRIC IDENTIFICATION WITH DEVICE IDENTIFICATION
    6.
    发明申请
    SUPPLEMENTING BIOMETRIC IDENTIFICATION WITH DEVICE IDENTIFICATION 有权
    用设备识别来补充生物识别

    公开(公告)号:US20160294825A1

    公开(公告)日:2016-10-06

    申请号:US15186630

    申请日:2016-06-20

    发明人: Nir Nice Eyal Krupka

    IPC分类号: H04L29/06

    摘要: A computer may identify an individual according to one or more biometrics based on various physiological aspects of the individual, such as metrics of various features of the face, gait, fingerprint, or voice of the individual. However, biometrics are often computationally intensive to compute, inaccurate, and unable to scale to identify an individual among a large set of known individuals. Therefore, the biometric identification of an individual may be supplemented by identifying one or more devices associated with the individual (e.g., a mobile phone, a vehicle driven by the individual, or an implanted medical device). When an individual is registered for identification, various device identifiers of devices associated with the individual may be stored along with the biometrics of the individual. Individuals may then be identified using both biometrics and detected device identifiers, thereby improving the efficiency, speed, accuracy, and scalability of the identification.

    摘要翻译: 计算机可以基于个体的各种生理方面根据一个或多个生物特征识别个体,例如个体的各种特征的度量,步态,指纹或个体的声音。 然而,生物识别通常对于计算,不精确和无法扩展以在一大群已知个体中识别个体而计算密集。 因此,可以通过识别与个人(例如,移动电话,由个体驱动的车辆或植入的医疗设备)相关联的一个或多个设备来补充个体的生物特征识别。 当个人被注册用于识别时,可以与个体的生物特征一起存储与个人相关联的设备的各种设备标识符。 然后可以使用生物特征和检测到的设备标识符来识别个体,从而提高识别的效率,速度,准确性和可扩展性。

    Voice identification enrollment
    7.
    发明授权

    公开(公告)号:US11152006B2

    公开(公告)日:2021-10-19

    申请号:US16020911

    申请日:2018-06-27

    摘要: Examples are disclosed that relate to voice identification enrollment. One example provides a method of voice identification enrollment comprising, during a meeting in which two or more human speakers speak at different times, determining whether one or more conditions of a protocol for sampling meeting audio used to establish human speaker voiceprints are satisfied, and in response to determining that the one or more conditions are satisfied, selecting a sample of meeting audio according to the protocol, the sample representing an utterance made by one of the human speakers. The method further comprises establishing, based at least on the sample, a voiceprint of the human speaker.