Mitsubishi Electric Research Laboratories, Inc.

Mitsubishi Electric Corporation

Chen Feng

Yuichi Taguchi

Srikumar Ramalingam

Vineet R Kamat

Haruyuki Iwama

Anthony Vetro

Ming-Yu Liu

Dong Tian

Tim Marks

Esra Cansizoglu

Teng-Yok Lee

Siheng Chen

Yong Xiao

Chieh-Chi Kao

Abhinav Kumar

Duanshun Li

Yong-Dian Jian

Wenxuan Mou

Xiaoming Liu

Khalid Yousif

Zhiding Yu

Maja Krivokuca

Robert Cohen

A controller for executing a task based on probabilistic image-based landmark localization, uses a neural network, which is trained to process images of objects of a type having a structured set of landmarks to produce a parametric probability distribution defined by values of parameters for a location of each landmark in each processed image. The controller submits the set of input images to the neural network to produce the values of the parameters that define the parametric probability distribution over the location of each landmark in the structured set of landmarks of each input image. Further, the controller determines, for each input image, a global landmark uncertainty for the image based on the parametric probability distributions of landmarks in the input image and executes the task based on the parametric probability distributions of landmarks in each input image and the global landmark uncertainty of each input image.

Image processing system and method for landmark location estimation with uncertainty

A first T-spline fitting system includes a three-dimensional (3D) sensor, a processor and a memory storing a program for representing a point cloud of the scene using T-spline representation. The system executes acquiring an image of the scene using the 3D sensor, generating the point cloud from the image, determining a parameterized domain of the point cloud by mapping data points of the point cloud to values of parameters of the parameterized domain, partitioning recursively the parameterized domain into a set of patches, connecting the patches to form a T-mesh defining topology of control points in the parameterized domain, and fitting a T-spline surface into the point cloud according to the T-mesh to determine positions of the control points.

Fast T-spline fitting system and method

A system for reconstructing a three-dimensional (3D) model of a scene including a point cloud having points identified by 3D coordinates includes at least one sensor to acquire a set of images of the scene from different poses defining viewpoints of the images and a memory to store the set of images and the 3D model of the scene. The system also includes a processor operatively connected to the memory and coupled with stored instructions to transform the images from the set of images to produce a set of virtual images of the scene viewed from virtual viewpoints; compare at least some features from the images and the virtual images to determine the viewpoint of each image in the set of images; and update 3D coordinates of at least one point in the model of the scene to match coordinates of intersections of ray back-projections from pixels of at least two images corresponding to the point according to the viewpoints of the two images.

System and method for virtually-augmented visual simultaneous localization and mapping

An image processing system includes a memory to store a classifier and a set of labeled images for training the classifier, wherein each labeled image is labeled as either a positive image that includes an object of a specific type or a negative image that does not include the object of the specific type, wherein the set of labeled images has a first ratio of the positive images to the negative images. The system includes an input interface to receive a set of input images, a processor to determine a second ratio of the positive images, to classify the input images into positive and negative images to produce a set of classified images, and to select a subset of the classified images having the second ratio of the positive images to the negative images, and an output interface to render the subset of the input images for labeling.

Deep active learning method for civil infrastructure defect detection

An image processing system includes an interface to transmit and receive data via a network, a processor connected to the interface, a memory storing an image processing program modules executable by the processor, wherein the image processing program causes the processor to perform operations. The operations include providing a point cloud of an image including objects into a segmentation network, segmenting point-wisely the point cloud into multiple classes of the objects and detecting boundaries of the objects using the segmentation network, wherein the segmentation network outputs a probability of associating primitive classes of the objects based on the segmented multiple classes and the segmented boundaries, verifying and refining the segmented multiple classes and the segmented boundaries using a predetermined fitting method, and correcting misclassification of the multiple classes of the objects by fitting the primitives to the multiple classes.

Method and apparatus for multi-model primitive fitting based on deep geometric boundary and instance aware segmentation

A image processing system for multi-label semantic edge detection in an image includes an image interface to receive an image of a scene including at least one object, a memory to store a neural network trained for performing a multi-label edge classification of input images assigning each pixel of edges of objects in the input images into one or multiple semantic classes, a processor to transform the image into a multi-label edge-map using the neural network detecting an edge of the object in the image and assigning multiple semantic labels to at least some pixels forming the edge, and an output interface to render the multi-label edge-map.

Multi-label semantic boundary detection system

A system controls motion of vehicle according to a first parking path ending at a target state and determines a second parking path from a current state to the target state using a model of the parking space. The system acquires images of the parking space during the motion of the vehicle along the first parking path and constructs the model of the parking space. The model is used to generate a set of virtual images of the environment of the parking space as viewed from virtual viewpoints outside of the first parking path. The current state of the vehicle is determined by comparing a current image of the parking space with at least one virtual image. The second parking path is determined from the current state to the target state.

Vehicle automated parking system and method

Systems and methods for determining a pattern in time series data representing an operation of a machine. A memory to store and provide a set of training data examples generated by a sensor of the machine, wherein each training data example represents an operation of the machine for a period of time ending with a failure of the machine. A processor configured to iteratively partition each training data example into a normal region and an abnormal region, determine a predictive pattern absent from the normal regions and present in each abnormal region only once, and determine a length of the abnormal region. Outputting the predictive pattern via an output interface in communication with the processor or storing the predictive pattern in memory, wherein the predictive pattern is a predictive estimate of an impending failure and assists in management of the machine.

Methods and systems for fast resampling method and apparatus for point cloud data

A method determines dimensions in a scene by first acquiring a depth image of the scene acquired by a sensor, and extracting planes from the depth image. Topological relationships of the planes are determined. The dimensions are determined based on the planes and the topological relationships. A quality of the dimensions is evaluated using a scene type, and if the quality is sufficient outputting the dimensions, and otherwise outputting a guidance to reposition the sensor.

Method for determining dimensions in an indoor scene from a single depth image

A method extracts planes from three-dimensional (3D) points by first partitioning the 3D points into disjoint regions. A graph of nodes and edges is then constructed, wherein the nodes represent the regions and the edges represent neighborhood relationships of the regions. Finally, agglomerative hierarchical clustering is applied to the graph to merge regions belonging to the same plane.

Method for extracting planes from 3D point cloud sensor data

Three-dimensional data are registered by selecting a first set of primitives from the data in a first coordinate system, wherein the first set of primitives includes at least one plane, at least one point, and a third primitive that is either a point or a plane, and selecting a second set of primitives from the data in a second coordinate system, wherein the second set of primitives includes at least one plane, at least one point, and a third primitive corresponding to the third primitive in the first set of primitives. Then, the planes are registered with each other, as are the points, to obtain registered primitives.

Growing community of inventors

Cambridge, MA, United States of America

Chen Feng