In an object detection device to be installed to a vehicle and detect an object outside the vehicle, a position calculator sets multiple candidate points representing a candidate position of the object, based on positions of feature points extracted from a first image captured at a first time. The multiple candidate points are set to be denser within a detection range set based on a distance to the object detected by the ultrasonic sensor than outside the detection range. The position calculator estimates positions of the multiple candidate points at a second time which is after the first time, based on the positions of the multiple candidate points and movement information of the vehicle, and calculates the position of the object by comparing the estimated positions of the multiple candidate points at the second time and the positions of the feature points extracted from a second image captured at the second time.

A method and apparatus with selective combined authentication performs a single authentication based on a first modality among plural modalities, and in response to the single authentication having failed, determines whether to perform a combined authentication by a combination of two or more of the plural modalities, and selectively, depending on a result of the determining of whether to perform the combined authentication, performs the combined authentication.

Methods and systems for explainable machine learning are described. In an example, a processor can receive a data set from a plurality of data sources corresponding to a plurality of domains. The processor can train a machine learning model to learn a plurality of vectors that indicate impact of the plurality of domains on a plurality of assets. The machine learning model can be operable to generate forecasts relating to performance metrics of the plurality of assets based on the plurality of vectors. In some examples, the machine learning model can be a neural attention network with shared hidden layers.

A cascaded machine learning inference system and method is disclosed. The cascaded system and method may be designed to be employed in resource restricted environments. The cascaded system and method may be applicable for applications that operate with limited power (e.g., a wearable smart watch). The cascaded system and method may employ two or more subsystems that are operable to classify an input signal provided by any number or types of sensors suitable for a given application. For instance, the sensors used may include gyroscopes, accelerometers, magnetometers, or barometric altimeters. The system and method may also be further split functionality across additional or new subsystems. By splitting operations and functionality across additional subsystems, the overall power consumption may further be reduced.

State of art techniques performing image labeling of remotely sensed data are computation intensive, consume time and resources. A method and system for efficient retrieval of a target in an image in a collection of remotely sensed data is disclosed. Image scanning is performed efficiently, wherein only a small percentage of pixels from the entire image are scanned to identify the target. One or more samples are intelligently identified based on sample selection criteria and are scanned for detecting presence of the target based on cumulative evidence score Plurality of sampling approaches comprising active sampling, distributed sampling and hybrid sampling are disclosed that either detect and localize the target or perform image labeling indicating only presence of the target.

Pixel image data of a scene is received in which the pixel image data includes a two-dimensional representation of an object in the scene. Point cloud data including three-dimensional point coordinates of a physical object within the scene corresponding to the two-dimensional representation of the object is received. The three-dimensional point coordinates include depth information of the physical object. The point cloud data is mapped to an image plane of the pixel image data to form integrated pixel image data wherein one or more pixels of the pixel image data have depth information integrated therewith. A three-dimensional bounding box is predicted for the object using a convolutional neural network based upon the integrated pixel image data.

Broadly speaking, the present application generally relates to a method for training a machine learning, ML, model to perform real world cross-modal retrieval problems, and to a computer-implemented method and apparatus for performing real world cross-modal retrieval problems such as including text-based video retrieval, sketch-based image retrieval, and image-text retrieval using a trained machine learning, ML, model.

A LIDAR system for use in a vehicle is provided. The LIDAR system may include at least one processor configured to control at least one light source for illuminating a field of view and scan a field of view by controlling movement of at least one deflector at which the at least one light source is directed. The at least one processor may also be configured to receive, from at least one sensor, reflections signals indicative of light reflected from an object in the field of view. The at least one processor may further be configured to detect at least one temporal distortion in the reflections signals, and determine from the at least one temporal distortion an angular orientation of at least a portion of the object.

A retrieval method includes obtaining first data corresponding to a retrieval object, wherein the first data indicates M feature vectors of the retrieval object, wherein each feature vector of the retrieval object corresponds to one modality of the retrieval object, and wherein M is an integer greater than 1, and obtaining a correlation between a plurality of groups of object information and the retrieval information to output at least one group of retrieved object information, wherein each group of object information corresponds to M feature vectors in an index, wherein the M feature vectors of each group of object information are indicated by one group of second data, and wherein each feature vector of the object information corresponds to one modality of the object information.

This document describes systems and techniques for clustering track pairs for multi-sensor track association. Many track-association algorithms use pattern-matching processes that can be computationally complex. Clustering tracks derived from different sensors present on a vehicle may reduce the computational complexity by reducing the pattern-matching problem into groups of subproblems. The weakest connection between two sets of tracks is identified based on both the perspective from each track derived from a first sensor and the perspective of each track derived from a second sensor. By identifying and pruning the weakest connection between two sets of tracks, a large cluster of tracks may be split into smaller clusters. The smaller clusters may require fewer computations by limiting the quantity of candidate track pairs to be evaluated. Fewer computations result in processing the sensor information more efficiently that, in turn, may increase the safety and reliability of an automobile.