Abstract: Systems, devices and methods provide, implement, and use vision-based methods of sequence inference for a device affixed to a vehicle.

Abstract: This invention classifies an action that appears in a video clip by receiving a video clip for analysis, applying a convolutional neural network mechanism (CNN) to the frames in the clip to generate a 4D embedding tensor for each frame in the clip, applying a multi-resolution convolutional neural network mechanism (CNN) to each of the frames in the clip to generate a sequence of reduced resolution blocks, computing a kinematic attention weight that estimates the amount of motion in the block, applying the attention weights to the embedding tensors for each frame in a clip, to generate a weighted embedding tensor, or context, that represents all the frames in the clip, at the resolution, combining the contexts across all resolutions to generate a multi-resolution context, performing a 3D pooling to obtain a 1D feature vector and classifying a primary action of the video clip based on the feature vector.

Abstract: An image sensor is positioned such that a field-of-view of the image sensor encompasses at least a portion of a rack storing items. The image sensor generates angled-view images of the items stored on the rack. A tracking subsystem determines that a person is within a threshold distance of the rack and receives image frames of the angled-view images. A pixel position of a wrist of the person is determined in at least a subset of the received image frames, thereby determining a set of pixel positions of the wrist. An aggregated wrist position is determined based on the set of pixel positions. If the aggregated wrist position is determined to correspond to a position on a shelf of the rack, a trigger signal is provided indicating a shelf-interaction event has occurred.

Abstract: Embodiments of the disclosure provide a disparity map acquisition method and apparatus, a device, a control system and a storage medium. The method includes: respectively performing feature extraction on left-view images and right-view images of a captured object layer by layer through M cascaded feature extraction layers, to obtain a left-view feature map set and a right-view feature map set of each layer, M being a positive integer greater than or equal to 2; constructing an initial disparity map based on the left-view feature map set and the right-view feature map set extracted by an Mth feature extraction layer; and iteratively refining, starting from an (M?1)th layer, the disparity map through the left-view feature map set and the right-view feature map set extracted by each feature extraction layer in sequence until a final disparity map is obtained based on an iteratively refined disparity map of a first layer.

Abstract: This application discloses a method and apparatus for object detection in an image, a vehicle, and a robot. The method for object detection in an image is performed by a computing device. The method includes determining an image feature of an image; determining a correlation of pixels in the image based on the image feature; updating the image feature of the image based on the correlation to obtain an updated image feature; and determining an object detection result in the image according to the updated image feature.

Abstract: Various embodiments include methods, apparatuses, systems, and computing devices that are configured to: (1) receive media of a processing region of an industrial process, where the processing region comprises at least one object, the media comprises a plurality of media elements, and each of the media elements comprises a field of view of the at least one object; (2) identify based on an area of interest, a set of pixels, wherein the field of view comprises the area of interest; (3) for each media element of the plurality of media elements: (a) extract an attribute value from each pixel of the set of pixels found in the media element, and (b) construct a respective array comprising each attribute value; (4) combine each of the respective arrays in a data structure; and (5) analyze the data structure to provide data on a processing parameter associated with the industrial process.

Abstract: The present disclosure provides techniques for extracting structural information using machine learning. One example method includes receiving electronic data indicating one or more pages, constructing, for each page of the one or more pages, a tree based on the page, wherein each level of the tree includes one or more nodes corresponding to elements in a level of elements in the page, encoding, for each page of the one or more pages, a value of each node of the tree for the page into a vector using a first machine learning model, sampling a plurality of pairs of vectors from the one or more trees for the one or more pages, wherein a given pair of vectors corresponds to values of nodes in a same tree, training a second machine learning model using the plurality of pairs, and combining each vector with weights of the second machine learning model.

Abstract: An electronic device and a method for controlling the electronic device, the method including receiving an input of a command to reproduce music contents, determining audio characteristics information on the music contents and situation information on an environment where the music contents are being reproduced; and displaying a visualization effect of visualizing the music contents using the audio characteristics information and the situation information, and reproducing the music contents.

Abstract: Disclosed are methods, devices, and computer-readable media for detecting lanes and objects in image frames of a monocular camera. In one embodiment, a method is disclosed comprising receiving a sample set of image frames; detecting a plurality of markers in the sample set of image frames using a convolutional neural network (CNN); fitting lines based on the plurality of markers; detecting a plurality of vanishing points based on the lines; identifying a best fitting horizon for the sample set of image frames via a RANSAC algorithm; computing an inverse perspective mapping (IPM) based on the best fitting horizon; and computing a lane width estimate based on the sample set of image frames using the IPM in a rectified view and the parallel line fitting.

Abstract: A system and method for estimating free space including applying a machine learning model to camera images of a navigation area, where the navigation area is broken into cells, synchronizing point cloud data from the navigation area with the processed camera images, and associating probabilities that the cell is occupied and object classifications of objects that could occupy the cells with cells in the navigation area based on sensor data, sensor noise, and the machine learning model.

Abstract: An Artificial Intelligence (AI) roof deterioration analysis system that tracks changes in roofs over time by evaluating a series of features in high-resolution images.

Abstract: Techniques related to training and implementing a bidirectional pairing architecture for object detection are discussed. Such techniques include generating a first enhanced feature map for each frame of a video sequence by processing the frames in a first direction, generating a second enhanced feature map for frame by processing the frames in a second direction opposite the first, and determining object detection information for each frame using the first and second enhanced feature map for the frame.

Abstract: Methods and systems for finding and ranking video sequences that contain tactical sequences (e.g., an uninterrupted portion of a sporting event) that are similar to tactical event content in a target video sequence, such as those occurring in sports, include learning and labelling (e.g., through use of metadata and the like) tactically significant sequences in a plurality of video segments, including a plurality of such tactically significant sequences in a single video segment.

Abstract: Systems and methods for tracking objects are disclosed herein. In one embodiment, a system having a processor merges features of detected objects extracted from a point cloud and a corresponding image to generate fused features for the detected objects, generates a learned distance metric for the detected objects using the fused features, determines matched detected objects and unmatched detected objects, applies prior tracking identifiers of the detected objects at the prior time to the matched detected objects, determines a confidence score for the fused features of the unmatched detected objects, and applies new tracking identifiers to the unmatched detected objects based on the confidence score.

Abstract: This disclosure relates to the use of “pseudo-images” to perform image recognition, e.g., to perform facial image recognition. In an embodiment, the pseudo-image is obtained by starting with a real world image and, after optional preprocessing, subjecting the image to a non-linear transformation that converts the image into a pseudo-image. While real world objects (or, more generally, real world patterns) may be perceivable in the starting image, they cannot be perceived in the pseudo-image. Image recognition takes place by comparing the pseudo-image with a library of known pseudo-images, i.e., image recognition takes place in pseudo-image space without a return to real world space. In this way, robust image recognition is achieved even for imperfect real world images, such as, real world images that have been degraded by noise, poor illumination, uneven lighting, and/or occlusion, e.g., the presence of glasses, scarves, or the like in the case of facial images.

Abstract: Methods and apparatus for the analysis of moving objects. In one embodiment, a pitch tracking system is disclosed that includes one or more cameras that have been positioned in a desired location and a computing system containing executable software that is configured to receive imaging data from the one or more cameras that captured the desired location and analyze seam placement and/or finger placement as a function of time for the received imaging data. Computer readable media and methodologies are also disclosed herein.

Abstract: This disclosure describes methods, non-transitory computer readable storage media, and systems that generate synthetized digital images via multi-resolution generator neural networks. The disclosed system extracts multi-resolution features from a scene representation to condition a spatial feature tensor and a latent code to modulate an output of a generator neural network. For example, the disclosed systems utilizes a base encoder of the generator neural network to generate a feature set from a semantic label map of a scene. The disclosed system then utilizes a bottom-up encoder to extract multi-resolution features and generate a latent code from the feature set. Furthermore, the disclosed system determines a spatial feature tensor by utilizing a top-down encoder to up-sample and aggregate the multi-resolution features. The disclosed system then utilizes a decoder to generate a synthesized digital image based on the spatial feature tensor and the latent code.

Abstract: A video quality evaluation method comprises acquiring an image sequence and audio information by decoding a to-be-evaluated video, wherein the to-be-evaluated video is non-offending video; extracting an action feature vector and a face feature vector from the image sequence, and extracting an audio feature vector from the audio information; constructing a video feature vector according to at least one of the action feature vector, the face feature vector and the audio feature vector; and determining a quality score of the to-be-evaluated video according to the video feature vector.

Abstract: Systems and methods for determining dwell time is provided. The method includes receiving images of an area including one or more people from one or more cameras, and detecting a presence of each of the one or more people in the received images using a worker. The method further includes receiving by the worker digital facial features stored in a watch list from a master controller, and performing facial recognition and monitoring the dwell time of each of the one or more people. The method further includes determining if each of the one or more people is in the watch list or has exceeded a dwell time threshold.

Abstract: The disclosure relates to system and method for providing assistance to a user using augmented reality. The method includes acquiring a video stream and a set of data associated with a task being performed by a user, in real-time, using a camera and/or a sensor device. The video stream includes sequential frames. The method further includes determining a present state associated with the task based on the sequential frames using an Artificial Neural Network (ANN) based action prediction model; determining scenarios and events corresponding to the scenarios based on the video stream and the set of data using an ANN based augmented intelligence model; and determining sequential instructions required for assisting the user to accomplish the task, dynamically, based on the present state and the events associated with the task, using at least one of a rule-based engine and an ANN based instruction prediction model.