Mitsubishi Electric Research Laboratories, Inc.

Mitsubishi Electric Corporation

Tim Marks

Michael J Jones

Anoop Cherian

Chiori Hori

Oncel C Tuzel

Alan Sullivan

Takaaki Hori

Yuichi Taguchi

Jonathan Le Roux

Fatih M Porikli

Hassan Mansour

John R Hershey

Ashok Veeraraghavan

Yudai Nakamura

Rohith Mv

Jie Ni

Ewa Nowara

Anthony Vetro

Matthew E Brand

Jay E Thornton

Ye Wang

Bret Harsham

Chen Feng

Bharat Singh

Siheng Chen

Ejaz Ahmed

Michael Jones

Goncalo José Dias Pais

Chavdar Papazov

Safa Medin

Ashok Veeraghavan

Kuldeep Kulkarni

Abhinav Kumar

Salil Tambe

Wenxuan Mou

Suren Kumar

Xiaoming Liu

Ritwik Kumar

An imaging controller is provided for segmenting instances from depth images including objects to be manipulated by a robot. The imaging controller includes an input interface configured to receive a depth image that includes objects, a memory configured to store instructions and a neural network trained to segment instances from the objects in the depth image, and a processor, coupled with the memory, configured to perform the instructions to segment a pickable instance using the trained neural network. The instructions include steps of selecting a tallest point in the depth image, defining a region using a shape such that the region surrounds the tallest point, sampling points in the region of the depth image, computing depth-geodesics between the tallest point and the sampled points, submitting the depth-geodesics to the neural network to segment the pickable instance among instances of the objects in the depth image, and an output interface configured to output a geometrical feature of the pickable instance to a manipulator controller of the robot.

Discriminative 3D shape modeling for few-shot instance segmentation

A remote photoplethysmography (RPPG) system for estimating vital signs of a person is provided. The RPPG system is configured to receive a set of imaging photoplethysmography (iPPG) signals measured from different regions of a skin of a person. The RPPG system is further configured to determine frequency coefficients at the frequency bins of the quantized frequency spectrum of the measured iPPG signals by minimizing a distance between the measured iPPG signals and corresponding iPPG signals reconstructed from the determined frequency coefficients, while enforcing joint sparsity of the determined frequency coefficients subject to the sparsity level constraint, such that the determined frequency coefficients of different iPPG signals have the non-zero values at the same frequency bins; and output one or a combination of the determined frequency coefficients, the iPPG signals reconstructed from the determined frequency coefficients, and a vital sign signal corresponding to the reconstructed iPPG signals.

System and method for remote measurements of vital signs of a person in a volatile environment

Present disclosure discloses an image processing system and method for manipulating two-dimensional (2D) images of three-dimensional (3D) objects of a predetermined class (e.g., human faces). A 2D input image of a 3D object of the predetermined class is manipulated by manipulating physical properties of the 3D object, such as a 3D shape of the 3D input object, an albedo of the 3D input object, a pose of the 3D input object, and lighting illuminating the 3D input object. The physical properties are extracted from the 2D input image using a neural network that is trained to reconstruct the 2D input image. The 2D input image is reconstructed by disentangling the physical properties from pixels of the 2D input image using multiple subnetworks. The disentangled physical properties produced by the multiple subnetworks are combined into a 2D output image using a differentiable renderer.

System and method for manipulating two-dimensional (2D) images of three-dimensional (3D) objects

System and method for generating verisimilar images from real depth images. Train a generative adversarial neural network (GAN) by accessing test depth images having identical instances as instances of a real depth image. Input the test depth images in the generator to generate estimated depth images representing an implicit three-dimensional model of the object. Input, each estimated depth image into a discriminator to obtain a loss and into a pose encoder to obtain a matching loss. Iteratively repeat processes until the losses are minimized to a threshold, to end training. Identify the instances in the real image using the trained GAN pose encoder, to produce a pose transformation matrix for each instance in the real image. Identify pixels in the depth images corresponding to the instances of the real image and merge the pixels for the depth images to form an instance segmentation map for the real depth image.

InSeGAN: a generative approach to instance segmentation in depth images

A navigation system for providing driving instructions to a driver of a vehicle traveling on a route is provided. The driving instructions are generated by executing a multimodal fusion method that comprises extracting features from sensor measurements, annotating the features with directions for the vehicle to follow the route with respect to objects sensed by the sensors, and encoding the annotated features with a multimodal attention neural network to produce encodings. The encodings are transformed into a common latent space, and the transformed encodings are fused using an attention mechanism producing an encoded representation of the scene. The method further comprises decoding the encoded representation with a sentence generation neural network to generate a driving instruction and submitting the driving instruction to an output device.

Method and system for scene-aware interaction

Embodiments of the present disclosure discloses a scene-aware video encoder system. The scene-aware encoder system transforms a sequence of video frames of a video of a scene into a spatio-temporal scene graph. The spatio-temporal scene graph includes nodes representing one or multiple static and dynamic objects in the scene. Each node of the spatio-temporal scene graph describes an appearance, a location, and/or a motion of each of the objects (static and dynamic objects) at different time instances. The nodes of the spatio-temporal scene graph are embedded into a latent space using a spatio-temporal transformer encoding different combinations of different nodes of the spatio-temporal scene graph corresponding to different spatio-temporal volumes of the scene. Each node of the different nodes encoded in each of the combinations is weighted with an attention score determined as a function of similarities of spatio-temporal locations of the different nodes in the combination.

Scene-aware video encoder system and method

A scene captioning system is provided. The scene captioning system includes an interface configured to acquire a stream of scene data signals including frames and sound data, a memory to store a computer-executable scene captioning model including a scene encoder, a timing decoder, a timing detector, and a caption decoder, wherein the audio-visual encoder is shared by the timing decoder and the timing detector and the caption decoder, and a processor, in connection with the memory. The processor is configured to perform steps of extracting scene features from the scene data signals by use of the audio-visual encoder, determining a timing of generating a caption by use of the timing detector, wherein the timing is arranged an early stage of the stream of scene data signals, and generating the caption based on the scene features by using the caption decoder according to the timing.

Low-latency captioning system

A computer-implemented method for training a dialogue response generation system and the dialogue response generation system are provided. The method includes arranging a first multimodal encoder-decoder for the dialogue response generation or video description having a first input and a first output, wherein the first multimodal encoder-decoder has been pretrained by training audio-video datasets with training video description sentences, arranging a second multimodal encoder-decoder for dialog response generation having a second input and a second output, providing first audio-visual datasets with first corresponding video description sentences to the first input of the first multimodal encoder-decoder, wherein the first encoder-decoder generates first output values based on the first audio-visual datasets with the first corresponding description sentences, providing the first audio-visual datasets excluding the first corresponding video description sentences to the second multimodal encoder-decoder. In this case, the second multimodal encoder-decoder generates second output values based on the first audio-visual datasets without the first corresponding video description sentences.

System and method for a dialogue response generation system

A remote photoplethysmography (RPPG) system includes an input interface to receive a sequence of measurements of intensities of different regions of a skin of a person indicative of vital signs of the person; a solver to solve an optimization problem to determine frequency coefficients of photoplethysmographic waveforms corresponding to the measured intensities at the different regions, wherein the solver determines the frequency coefficients to reduce a distance between intensities of the skin reconstructed from the frequency coefficients and the corresponding measured intensities of the skin while enforcing joint sparsity on the frequency coefficients; and an estimator to estimate the vital signs of the person from the determined frequency coefficients of photoplethysmographic waveforms.

System and method for remote measurements of vital signs

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 method and system determine a three-dimensional (3D) pose of an object and 3D locations of landmark points of the object by first obtaining a 3D point cloud of the object. 3D surface patches are extracted from the 3D point cloud, and a parametric model is fitted to each 3D surface patch to determine a set of descriptors. A set of correspondences between the set of descriptors and a set of descriptors of patches extracted from 3D point clouds of objects from the same object class with known 3D poses and known 3D locations of landmark points is determined. Then, the 3D pose of the object and 3D locations of the landmark points of the object are estimated from the set of correspondences.

Method and system for determining 3D object poses and landmark points using surface patches

A system for generating a word sequence includes one or more processors in connection with a memory and one or more storage devices storing instructions causing operations that include receiving first and second input vectors, extracting first and second feature vectors, estimating a first set of weights and a second set of weights, calculating a first content vector from the first set of weights and the first feature vectors, and calculating a second content vector, transforming the first content vector into a first modal content vector having a predetermined dimension and transforming the second content vector into a second modal content vector having the predetermined dimension, estimating a set of modal attention weights, generating a weighted content vector having the predetermined dimension from the set of modal attention weights and the first and second modal content vectors, and generating a predicted word using the sequence generator.

Method and system for multi-modal fusion model

A method and system detects actions of an object in a scene by first acquiring a video of the scene as a sequence of images, wherein each image includes pixels, wherein the video is partitioned into chunks. The object in the video is tracked. For each object and each chunk of the video, trajectories of the pixels within a bounding box located over the object are tracked, and cropped trajectories and cropped images for one or more images in the chunk are produced using the bounding box. Then, the cropped trajectories and cropped images are passed to a recurrent neural network (RNN) that outputs a relative score for each action of interest.

Method and system for detecting actions in videos

A method and system for detecting actions of an object in a scene from a video of the scene. The video is a video sequence partitioned into chunks, and each chunk includes consecutive video frames. The method including the following elements. Acquiring the video of the scene, wherein the video includes a sequence of images. Tracking the object in the video, and for each object and each chunk of the video, further comprising: determining a sequence of contour images from video frames of the video sequence to represent motion data within a bounding box located around the object. Using the bounding box to produce cropped contour images and cropped images for one or more images in each chunk. Passing the cropped contour images and the cropped images to a recurrent neural network (RNN) that outputs a relative score for each action of interest.

Method and system for detecting actions in videos using contour sequences

A method for face alignment operates on a face image and a set of intitial landmark locations by first aligning globally the initial locations to a set of landmark locations of a face with a prototype shape to obtain global alignment parameters, and then warping the initial locations and the image from a coordinate frame of the image to a coordinate frame of the prototype shape according to the global alignment parameters to obtain warped landmark locations and a warped face image. Features are extracted from the warped face image at the warped landmark locations, and a regression function is applied to the features to obtain updated landmark locations in the coordinate frame of the prototype shape. Finally, the updated landmark locations in the coordinate frame of the prototype shape are warped to the coordinate frame of the image, to obtain updated landmark locations.

Growing community of inventors

Newton, MA, United States of America

Tim Marks