Deep learning based method and apparatus for three dimensional model triangular facet feature learning and classifying

    公开(公告)号:US10049299B2

    公开(公告)日:2018-08-14

    申请号:US15439896

    申请日:2017-02-22

    摘要: The invention discloses a deep learning based method for three dimensional (3D) model triangular facet feature learning and classifying and an apparatus. The method includes: constructing a deep convolutional neural network (CNN) feature learning model; training the deep CNN feature learning model; extracting a feature from, and constructing a feature vector for, a 3D model triangular facet having no class label, and reconstructing a feature in the constructed feature vector using a bag-of-words algorithm; determining an output feature corresponding to the 3D model triangular facet having no class label according to the trained deep CNN feature learning model and an initial feature corresponding to the 3D model triangular facet having no class label; and performing classification. The method enhances the capability to describe 3D model triangular facets, thereby ensuring the accuracy of 3D model triangular facet feature learning and classifying results.

    METHOD FOR REAL-TIME CUTTING OF DIGITAL ORGAN BASED ON METABALL MODEL AND HYBRID DRIVING METHOD

    公开(公告)号:US20180204378A1

    公开(公告)日:2018-07-19

    申请号:US15835408

    申请日:2017-12-07

    摘要: Disclosed is a method for real-time cutting of digital organ based on a metaball model and a hybrid driving method, including a cutting procedure for driving a model using a position-based dynamics and a meshless method, a cutting mode which begins from a metaball driven by the position-based dynamics, proceeds to a point set driven by the meshless method and then create a new metaball. The method includes: a preprocessing procedure which performs an initialization operation while reading a model file; a deforming procedure which drives a model using a method based on the position-based dynamics; a cutting procedure which drives the model using the hybrid driving method and performs cutting using said cutting mode; and a rendering procedure which renders the model during the second and third procedures.

    High-density magnetic memory device

    公开(公告)号:US10020044B2

    公开(公告)日:2018-07-10

    申请号:US15795223

    申请日:2017-10-26

    摘要: A high-density magnetic memory device includes: a heavy metal strip or an antiferromagnet strip with a thickness of 0-20 nm, and a plurality of magnetic tunnel junctions manufactured thereon, wherein each of the magnetic tunnel junctions represents a memory bit, which from bottom to top comprises a first ferromagnetic metal with a thickness of 0-3 nm, an oxide with a thickness of 0-2 nm, a second ferromagnetic metal with a thickness of 0-3 nm, a synthetic antiferromagnetic layer with a thickness of 10-20 nm and a No. X top electrode with a thickness of 10-200 nm, wherein an X value is a serial number of the memory bit; two ends of the heavy metal strip or the antiferromagnet strip are respectively plated with a first bottom electrode and a second bottom electrode. The write operation for the memory device of the present invention is accomplished by applying unidirectional write currents.

    METHOD FOR SKINNING TECHNOLOGY BASED ON EXTENDED POSITION BASED DYNAMICS AND FOR WEIGHT RETARGETING IN CHARACTER ANIMATION

    公开(公告)号:US20180182152A1

    公开(公告)日:2018-06-28

    申请号:US15711954

    申请日:2017-09-21

    IPC分类号: G06T13/40 G06T17/20

    摘要: Disclosed relates to a method for skinning technology based on extended PBD and for weight retargeting in character animation. A corresponding tetrahedral mesh is generated for a given triangular mesh; a tensile constraint, a volume constraint, a self-collision constraint and an energy constraint of the extended PBD are initialized for the tetrahedral mesh, and vertex weights of the triangular mesh and the tetrahedral mesh are optimized simultaneously; the deformations of the triangular and the tetrahedral meshes are implemented through LBS technique; deformation results of the LBS are optimized through extended PBD; the joint region of the triangular mesh is smoothed through a Laplacian smoothing; bi-harmonic distance fields of the original mesh and a new mesh are calculated respectively, and two corresponding vertices of the mesh are given and a prediction function is defined; A weight retargeting algorithm is implemented in the present application and has a good practicability.

    High-density magnetic memory device
    85.
    发明申请

    公开(公告)号:US20180061482A1

    公开(公告)日:2018-03-01

    申请号:US15795223

    申请日:2017-10-26

    摘要: A high-density magnetic memory device includes: a heavy metal strip or an antiferromagnet strip with a thickness of 0-20 nm, and a plurality of magnetic tunnel junctions manufactured thereon, wherein each of the magnetic tunnel junctions represents a memory bit, which from bottom to top comprises a first ferromagnetic metal with a thickness of 0-3 nm, an oxide with a thickness of 0-2 nm, a second ferromagnetic metal with a thickness of 0-3 nm, a synthetic antiferromagnetic layer with a thickness of 10-20 nm and a No. X top electrode with a thickness of 10-200 nm, wherein an X value is a serial number of the memory bit; two ends of the heavy metal strip or the antiferromagnet strip are respectively plated with a first bottom electrode and a second bottom electrode. The write operation for the memory device of the present invention is accomplished by applying unidirectional write currents.

    KIND OF LOW MAGNETIC SENSITIVITY PM-PCF BASED ON MECHANICAL BUFFER

    公开(公告)号:US20180003889A1

    公开(公告)日:2018-01-04

    申请号:US15630155

    申请日:2017-06-22

    IPC分类号: G02B6/024 G02B6/02 G01C19/72

    CPC分类号: G02B6/024 G02B6/02314

    摘要: The low magnetic sensitivity PM-PCF based on mechanical buffer is obtained by adding buffer structures in the cladding layer of the photonic crystal fiber. In the center of the fiber, the core region contains at least 3 layers of air-holes, enclosed by the cladding layer. The buffer structures are placed in the cladding layer. These buffer structures are formed by replacing silica of any shape by air, and are symmetrically located in X-axis and Y-axis directions to achieve mechanical isotropy. The buffer structures improve the fiber's performance in fiber coiling and stress conditions. Therefore, the fiber optic gyroscope using the PM-PCF can do without a magnetic shield, thus greatly reducing the weight of the fiber optic gyroscope and extending the scope of its application. Compared with the conventional commercial PCF, the PM-PCF provides the fiber optic gyroscope with lower temperature sensitivity and improved accuracy.

    Method and system for anti-disturbance composite on-line guidance for atmosphere-entering phase of a Mars lander

    公开(公告)号:US09671789B1

    公开(公告)日:2017-06-06

    申请号:US15371506

    申请日:2016-12-07

    IPC分类号: B64G1/36 G05D1/08 B64G1/24

    摘要: A method and system for anti-interference compound on-line guidance in the atmosphere-entering stage of a Mars lander is provided in the present invention. The method comprises steps of a) building a dynamics model for the atmosphere-entering phase of a Mars lander to incorporating the disturbance brought by the Mars atmosphere density uncertainty into the dynamics model for the atmosphere-entering stage of the Mars lander; b) constructing a disturbance observer to estimate the disturbance brought by the Mars atmosphere density uncertainty in the dynamics model for the atmosphere-entering phase of the Mars lander; c) building a prediction-correction guidance law, and guiding the heeling angle amplitude of the prediction-correction guidance law by continuously updating undetermined parameters; d) constructing an anti-disturbance composite guidance law to compensate the Mars atmosphere density uncertainty; and e) adjusting the deviation of the landing point of the Mars lander by compensating the Mars atmosphere uncertainty disturbance.

    Method of sparse representation of contents of high-resolution video images supporting content editing and propagation
    89.
    发明授权
    Method of sparse representation of contents of high-resolution video images supporting content editing and propagation 有权
    支持内容编辑和传播的高分辨率视频图像内容的稀疏表示方法

    公开(公告)号:US09384781B2

    公开(公告)日:2016-07-05

    申请号:US14332813

    申请日:2014-07-16

    摘要: This invention provides a method of sparse representation of contents of high-resolution video images which supports content editing and propagation. It mainly comprises five steps which are: (1) to input the original images or videos of high resolution to summarize the characteristic information of their pixels; (2) to acquire the highly sparse samples of the original images or videos through the sparse representation technique; (3) to reconstruct each pixel of the input images or videos with a small number of the original sparse samples to calculate a coefficient of reconstruction; (4) to edit and propagate the original sparse samples to yield a result of new sparse samples; (5) to generate a result of final images or videos of high resolution according to the result of sparse samples and the coefficient of reconstruction. This invention only edits and propagates the highly sparse samples rather than all the information of pixels. Thus, the memory consumption can be greatly reduced so as to possibly process the images or videos of high resolution in a very small memory space. It is very potential to be widely applied in the fields of image processing, computer vision and augmented reality technique.

    摘要翻译: 本发明提供了一种支持内容编辑和传播的高分辨率视频图像内容的稀疏表示方法。 它主要包括五个步骤:(1)输入高分辨率的原始图像或视频,以总结其像素的特征信息; (2)通过稀疏表示技术获取原始图像或视频的高度稀疏样本; (3)用少量的原始稀疏样本重建输入图像或视频的每个像素,以计算重建系数; (4)编辑和传播原始稀疏样本以产生新的稀疏样本的结果; (5)根据稀疏样本和重建系数的结果生成最终图像或高分辨率视频的结果。 本发明仅编辑和传播高度稀疏的样本,而不是像素的全部信息。 因此,可以大大减少存储器消耗,以便可能在非常小的存储空间中处理高分辨率的图像或视频。 在图像处理,计算机视觉和增强现实技术领域中广泛应用是非常有潜力的。