Inventor

YANG MING-HSUAN

US35 patents

⚠️ This page may combine multiple inventors who share the name “YANG MING-HSUAN”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

YANG MING-HSUAN

HONDA MOTOR CO LTD

Monocular tracking of 3D human motion with a coordinated mixture of factor analyzers

Face recognition using kernel fisherfaces

Human pose estimation with data driven belief propagation

Simultaneous localization and mapping using multiple view feature descriptors

Direct method for modeling non-rigid motion with thin plate spline transformation

Face recognition system

Clustering appearances of objects under varying illumination conditions

Discriminative motion modeling for human motion tracking

Fast human pose estimation using appearance and motion via multi-dimensional boosting regression

Visual tracking using incremental fisher discriminant analysis

Detecting humans via their pose

Adaptive probabilistic visual tracking with incremental subspace update

Extended Isomap using Fisher Linear Discriminant and Kernel Fisher Linear Discriminant

Video-based face recognition using probabilistic appearance manifolds

Adaptive discriminative generative model and application to visual tracking

Method, apparatus and program for detecting an object

System and method for providing unsupervised domain adaptation for spatio-temporal action localization

System and method for providing unsupervised domain adaptation for spatio-temporal action localization

Image clustering with metric, local linear structure, and affine symmetry

NVIDIA CORP

Bilateral convolution layer network for processing point clouds

Photorealistic image stylization using a neural network model

Training a neural network to predict superpixels using segmentation-aware affinity loss

Using residual video data resulting from a compression of original video data to improve a decompression of the original video data

Learning affinity via a spatial propagation neural network

Using residual video data resulting from a compression of original video data to improve a decompression of the original video data

Training a neural network to predict superpixels using segmentation-aware affinity loss

GOOGLE LLC

Resource constrained neural network architecture search

Real time perspective correction on faces

Learnable cost volume for determining pixel correspondence

Real time perspective correction on faces

Compression-informed video super-resolution

YANG MING-HSUAN

Online sparse matrix Gaussian process regression and visual applications

Online articulate object tracking with appearance and shape

MITSUBISHI ELECTRIC RES LABS I

Gender classification with support vector machines

INST INFORMATION IND

Abnormal packet detection apparatus and method