Presented are intelligent vehicle systems with networked on-body vehicle cameras with camera-view augmentation capabilities, methods for making/using such systems, and vehicles equipped with such systems. A method for operating a motor vehicle includes a system controller receiving, from a network of vehicle-mounted cameras, camera image data containing a target object from a perspective of one or more cameras. The controller analyzes the camera image to identify characteristics of the target object and classify these characteristics to a corresponding model collection set associated with the type of target object. The controller then identifies a 3D object model assigned to the model collection set associated with the target object type. A new “virtual” image is generated by replacing the target object with the 3D object model positioned in a new orientation. The controller commands a resident vehicle system to execute a control operation using the new image.

Example methods, apparatus, and articles of manufacture to classify labels based on images using artificial intelligence are disclosed. An example apparatus includes a regional proposal network to determine a first bounding box for a first region of interest in a first input image of a product; and determine a second bounding box for a second region of interest in a second input image of the product; a neural network to: generate a first classification for a first label in the first input image using the first bounding box; and generate a second classification for a second label in the second input image using the second bounding box; a comparator to determine that the first input image and the second input image correspond to a same product; and a report generator to link the first classification and the second classification to the product.

In one aspect, there is provided a method performed by one or more data processing apparatus, the method including receiving a training dataset having multiple training examples, where each training example includes: (i) an image, and (ii) a segmentation defining a target region of the image that has been classified as including pixels in a target category. The method further includes determining a respective refined segmentation for each training example, including, for each training example, processing the target region of the image defined by the segmentation for the training example using a de-noising neural network to generate a network output that defines the refined segmentation for the training example. The method further includes training a segmentation machine learning model on the training examples of the training dataset, including, for each training example training the segmentation machine learning model to process the image included in the training example to generate a model output that matches the refined segmentation for the training example.

A system, method, and computer program product for implementing video action recognition is provided. The method includes receiving a video stream comprising user movement actions. Skeleton points associated with a video representation of a user executing the user movement actions are extracted and categorized with respect to multiple digital levels. Initial visual windows points are generated within video frames and an average movement distance for the group of skeleton points are determined with respect to the video frames. In response, sizes for the visual windows are adjusted and feature vectors are extracted from the group of skeleton points. Point coordinates of the skeleton points are extracted and linked with the feature vectors. A convolutional neural network associated with linking the feature vectors with the point coordinates is generated and the video stream is enabled with respect to video action recognition associated with accurate presentation of the video stream.

Method and system for assisting electronic chip design, comprising: receiving netlist data for a proposed electronic chip design, the netlist data including a list of circuit elements and a list of interconnections between the circuit elements; converting the netlist data to a graph that represents at least some of the circuit elements as nodes and represents the interconnections between the circuit elements as edges; extracting network embeddings for the nodes based on a graph topology represented by the edges; extracting degree features for the nodes based on the graph topology; and computing, using a graph neural network, a congestion prediction for the circuit elements that are represented as nodes based on the extracted network embeddings and the extracted degree features.

The present disclosure describes techniques for, in addition to a “global model” that is trained using training data, training one or more “segmented” machine learning (ML) models using subsets or segments of the training data, wherein a segmented ML model is trained using a segment or subset of the training data that satisfies a particular condition. Both the trained global model, and where applicable, one or more trained segmented models are used to make predictions in a production environment.

Disclosed are a weight data storage method and a convolution computation method that may be implemented in a neural network. The weight data storage method comprises searching for effective weights in a weight convolution kernel matrix and acquiring an index of effective weights. The effective weights are non-zero weights, and the index of effective weights is used to mark the position of the effective weights in the weight convolution kernel matrix. The weight data storage method further comprises storing the effective weights and the index of effective weights. According to the weight data storage method and the convolution computation method of the present disclosure, storage space can be saved, and computation efficiency can be improved.

A method and system of graph feature extraction and graph classification based on adjacency matrix is provided. The invention first concentrates the connection information elements in the adjacency matrix into a specific diagonal region of the adjacency matrix which reduces the non-connection information elements in advance. Then the subgraph structure of the graph is further extracted along the diagonal direction using the filter matrix. Further, it uses a stacked convolutional neural network to extract a larger subgraph structure. On one hand, it greatly reduces the amount of computation and complexity, getting rid of the limitations caused by computational complexity and window size. On the other hand, it can capture large subgraph structure through a small window, as well as deep features from the implicit correlation structures at both vertex and edge level, which improves speed and accuracy of graph classification.

A data labeling method, apparatus and system are provided. The method includes: sampling a data source according to an evaluation task for the data source to obtain sampled data; generating a labeling task from the sampled data; sending the labeling task to a labeling device; and receiving a labeled result of the labeling task from the labeling device. As such, an automatic evaluation of data can be implemented by using the evaluation task, and evaluation efficiency is improved.

A quantization method and device for a neural network model, and a computer-readable storage medium are provided. The method includes determining, from a neural network model, a target convolution kernel having an abnormal coefficient distribution, splitting the target convolution kernel so as to obtain a plurality of sub-convolution kernels, quantizing the plurality of sub-convolution kernels respectively to obtain a plurality of quantized convolution kernels, and replacing the target convolution kernel with the plurality of quantized convolution kernels. The method can reduce quantization errors.