Abstract: A system and method for performing intersection scenario retrieval that includes receiving a video stream of a surrounding environment of an ego vehicle. The system and method also include analyzing the video stream to trim the video stream into video clips of an intersection scene associated with the travel of the ego vehicle. The system and method additionally include annotating the ego vehicle, dynamic objects, and their motion paths that are included within the intersection scene with action units that describe an intersection scenario. The system and method further include retrieving at least one intersection scenario based on a query of an electronic dataset that stores a combination of action units to operably control a presentation of at least one intersection scenario video clip that includes the at least one intersection scenario.

Abstract: A system and method for predicting driving actions based on intent-aware driving models that include receiving at least one image of a driving scene of an ego vehicle. The system and method also include analyzing the at least one image to detect and track dynamic objects located within the driving scene and to detect and identify driving scene characteristics associated with the driving scene and processing an ego-thing graph associated with the dynamic objects and an ego-stuff graph associated with the driving scene characteristics. The system and method further include predicting a driver stimulus action based on a fusion of representations of the ego-thing graph and the ego-stuff graph and a driver intention action based on an intention representation associated with driving intentions of a driver of the ego vehicle.

Abstract: A system and method for risk object identification via causal inference that includes receiving at least one image of a driving scene of an ego vehicle and analyzing the at least one image to detect and track dynamic objects within the driving scene of the ego vehicle. The system and method also include implementing a mask to remove each of the dynamic objects captured within the at least one image. The system and method further include analyzing a level of change associated with a driving behavior with respect to a removal of each of the dynamic objects. At least one dynamic object is identified as a risk object that has a highest level of influence with respect to the driving behavior.

Abstract: A system and method for learning and executing naturalistic driving behavior that include classifying a driving maneuver as a goal-oriented action or a stimulus-driven action based on data associated with a trip of a vehicle. The system and method also include determining a cause associated with the driving maneuver classified as a stimulus-driven action and determining an attention capturing traffic related object associated with the driving maneuver. The system and method additionally include building a naturalistic driving behavior data set that includes at least one of: an annotation of the driving maneuver based on a classification of the driving maneuver, an annotation of the cause, and an annotation of the attention capturing traffic object. The system and method further include controlling the vehicle to be autonomously driven based on the naturalistic driving behavior data set.

Abstract: Systems and methods for driver behavior recognition is provided. In one embodiment a computer implemented method includes receiving image data associated with a general objects. The method also includes identifying a reactive object and an inert object from the general objects based on the image data. An ego reactive graph is generated for the reactive object based on a reactive feature of the reactive object and a reactive position vector. An ego inert graph is generated for the inert object based on an inert feature of the inert object and an inert distance. The method further includes performing interaction modeling based on the ego reactive graphs and the ego inert graphs to generate updated features. The method also includes performing temporal modeling on the updated features. The method further includes determining an egocentric representation of a tactical driver behavior based at least in part on the updated features.

Abstract: An antenna structure includes a housing, a feeding portion, and a connecting portion. The housing defines a gap and a groove. The housing forms a radiating portion and a coupling portion through the gap and the groove. A portion of the housing between the feeding portion and the gap forms a first radiating section. The connecting portion is electrically connected to one end of the coupling portion adjacent to the gap. When the feeding portion supplies current, the current flows through the feeding portion and the first radiating section, and is coupled to the connecting portion through the gap to activate a first operating mode. When the feeding portion supplies current, the current flows through the feeding portion and the first radiating section, and is coupled to the coupling portion through the gap to activate a second operating mode.

Abstract: Multi-modal data representing driving events and corresponding actions related to the driving events can be obtained and used to train a neural network at least in part by using a triplet loss computed for the driving events as a regression loss to determine an embedding of driving event data. In some cases, using the trained neural network, a retrieval request for an input driving event and corresponding action can be processed by determining, from the neural network, one or more similar driving events or corresponding actions in the multi-modal data.

Abstract: An antenna structure includes a housing, a first feed source, a ground portion, a radiator, and a second feed source. The housing includes a front frame, a backboard, and a side frame. The side frame defines a slot. The front frame defines a first gap, a second gap, and a groove. A radiating portion and a coupling portion are divided from the housing by the slot, the first gap, the second gap, and the groove. The first feed source is electrically connected to the radiating portion. One end of the ground portion is electrically connected to the radiating portion and another end of the ground portion is grounded. The radiator is coupled with and apart from the coupling portion. The second feed source is electrically connected to the radiator and a current from the second feed source is coupled to the coupling portion through the radiator.

Abstract: A method and apparatus for minimizing silicon germanium facets in planar metal oxide semiconductor structures is disclosed. For example, a device fabricated according to the method may include a semiconductor substrate, a plurality of gate stacks formed on the substrate, a plurality of source/drain regions formed from silicon germanium, and a shallow trench isolation region positioned between two source/drain regions of the plurality of source/drain regions. Each source/drain region of the plurality of source/drain regions is positioned adjacent to at least one gate stack of the plurality of gate stacks. Moreover, the shallow trench isolation region forms a trench in the substrate without intersecting the two source/drain regions.

Abstract: A system and method for providing unsupervised domain adaption for spatio-temporal action localization that includes receiving video data associated with a surrounding environment of a vehicle. The system and method also include completing an action localization model to model a temporal context of actions occurring within the surrounding environment of the vehicle based on the video data and completing an action adaption model to localize individuals and their actions and to classify the actions based on the video data. The system and method further include combining losses from the action localization model and the action adaption model to complete spatio-temporal action localization of individuals and actions that occur within the surrounding environment of the vehicle.

Abstract: Methods, systems, and computer-readable mediums storing computer executable code for visual recognition implementing a triplet loss function are provided. The method include receiving an image generated from an image source associated with a vehicle. The method may also include analyzing the image based on a convolutional neural network. The convolutional neural network may apply both a triplet loss function and a softmax loss function to the image to determine classification logits. The method may also include classifying the image into a predetermined class distribution based upon the determined classification logits. The method may also include instructing the vehicle to perform a specific task based upon the classified image.

Abstract: The present disclosure generally relates to methods and systems for identifying objects from a 3D point cloud and a 2D image. The method may include determining a first set of 3D proposals using Euclidean clustering on the 3D point cloud and determining a second set of 3D proposals from the 3D point cloud based on a 3D convolutional neural network. The method may include pooling the first and second sets of 3D proposals to determine a set of 3D candidates. The method may include projecting the first set of 3D proposals onto the 2D image and determining a first set of 2D proposals using 2D convolutional neural network. The method may include pooling the projected first set of 3D proposals and the first set of 2D proposals to determine a set of 2D candidates then pooling the set of 3D candidates and the set of 2D candidates.

Abstract: Aspects of the present disclosure may include methods, apparatuses, and computer readable media for receiving one or more images having a plurality of objects, receiving a notification from an occupant of the self-driving vehicle, generating an attention map highlighting the plurality of objects based on at least one of the one or more images and the notification, and providing at least one of a steering control or a velocity control to operate the self-driving vehicle based on the attention map and the notification.

Abstract: Driver behavior recognition or driver behavior prediction are described herein. A first image sequence including image frames associated with a forward-facing image capture device of a vehicle and a corresponding vehicle data signal sequence may be received. A second image sequence including image frames associated with a rear or driver facing image capture device of the vehicle may be received. Feature vectors may be generated for respective sequences using neural networks, such as a convolutional neural network (CNN), a depth CNN, a recurrent neural network (RNN), a fully connected layer, a long short term memory (LSTM) layer, etc. A fusion feature may be generated by performing data fusion on any combination of the feature vectors. A predicted driver behavior may be generated based on the LSTM layer and n image frames on an image sequence and include x number of prediction frames.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for jointly training an image embedding model and a text embedding model. In one aspect, a method comprises: processing data from a historical query log of a search system to generate a candidate set of training examples, wherein each training example comprises: (i) a search query comprising a sequence of one or more words, (ii) an image, and (iii) selection data characterizing how often users selected the image in response to the image being identified by a search result for the search query; selecting a plurality of training examples from the candidate set of training examples; and using the training data to jointly train the image embedding model and the text embedding model.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an image embedding model. In one aspect, a method comprises: obtaining training data comprising a plurality of training examples, wherein each training example comprises: an image pair comprising a first image and a second image; and selection data indicating one or more of: (i) a co-click rate of the image pair, and (ii) a similar-image click rate of the image pair; and using the training data to train an image embedding model having a plurality of image embedding model parameters.

Abstract: Methods, systems, and computer-readable mediums storing computer executable code for visual recognition implementing a triplet loss function are provided. The method include receiving an image generated from an image source associated with a vehicle. The method may also include analyzing the image based on a convolutional neural network. The convolutional neural network may apply both a triplet loss function and a softmax loss function to the image to determine classification logits. The method may also include classifying the image into a predetermined class distribution based upon the determined classification logits. The method may also include instructing the vehicle to perform a specific task based upon the classified image.

Abstract: Multi-modal data representing driving events and corresponding actions related to the driving events can be obtained and used to train a neural network at least in part by using a triplet loss computed for the driving events as a regression loss to determine an embedding of driving event data. In some cases, using the trained neural network, a retrieval request for an input driving event and corresponding action can be processed by determining, from the neural network, one or more similar driving events or corresponding actions in the multi-modal data.

Abstract: A hydrogen generation method including steps as follows: adding a nitrogen-substituted derivative of an alkyldithiolate ruthenium complex as a biomimetic hydrogenase photocatalyst into a solution, adding an organic acid into the solution, adding a P-ligand into the solution, adding an electron donor into the solution, and irradiating the solution with light in order to generate hydrogen.

Abstract: A method of forming a layout of a semiconductor device includes the following steps. First line patterns extend along a first direction in a first area and a second area, but the first line patterns extend along a second direction in a boundary area. Second line patterns extend along a third direction in the first area and the second area, but the second line patterns extend along a fourth direction in the boundary area, so that minimum distances between overlapping areas of the first line patterns and the second line patterns in the boundary area are larger than minimum distances between overlapping areas of the first line patterns and the second line patterns in the first area and the second area. A trimming process is performed to shade the first line patterns and the second line patterns in the boundary area and the second area.