Robert Boschgmbh

Dimitrios Bariamis

Masato Takami

Uwe Brosch

Ferran Diego Andilla

Frank Moesle

Paul Ruhnau

Daniela Weiss

Emil Schreiber

A method for training an artificial neural network (ANN) capable of multitasking. A first path is provided for a first information flow through the ANN, the first path coupling an input layer with at least one task-spanning intermediate layer which is common to a plurality of differing tasks of the ANN. The first path couples the at least one task-specific intermediate layer with a respective task-specific ANN segment of the tasks. First training data are supplied via the input layer and the first path for training task-spanning parameters that are common to the tasks. At least one task-specific second path is provided for a second information flow through the ANN that differs from the first information flow. The second path couples the input layer with only a subset of the task-specific ANN segments of the tasks, and second training data are supplied for training task-specific parameters via the second path.

Method for training and operating an artificial neural network capable of multitasking, artificial neural network capable of multitasking, and device

A method for the computing and memory resource-conserving semantic segmentation of image data of an imaging sensor with the aid of an artificial neural network, in particular, a convolutional neural network, the artificial neural network including an encoder path and a decoder path. The method includes: dividing an input tensor into at least one first slice tensor and at least one second slice tensor as a function of a division function, the input tensor being dependent on the image data; outputting the at least one first slice tensor to the decoder path of the neural network; connecting the at least one first slice tensor to the at least one second slice tensor as a function of a connecting function to obtain an output tensor; and outputting the output tensor to the encoder path of the artificial neural network.

Method, artificial neural network, device, computer program, and machine-readable memory medium for the semantic segmentation of image data

An evaluation system for processing measured data which include physical measured data detected with the aid of one or multiple sensors, and/or realistic synthetic measured data of the sensor(s), into one or multiple evaluation results. The system includes at least two input stages independent from each other, which are designed to receive measured data and process these measured data into precursors. At least one processing stage, receives the precursors from all input stages as inputs and is designed to process one or multiple input precursor(s) into a shared intermediate product. At least one output stage, which is designed to process the intermediate product into one or multiple evaluation result(s) of the evaluation system. A method for training the evaluation system. A method for operating the evaluation system is also provided.

Evaluation system for measured data from multiple domains

Method for the calculation resource-saving semantic segmentation of image data of an imaging sensor with the aid of an artificial neural network, in particular, of a convolutional neural network, the artificial neural network including an encoder path, a decoder path, the encoder path transitioning into the decoder path, the transition taking place via a discriminative path, the following steps taking place in the discriminative path: dividing an input tensor as a function of a division function into at least one first slice tensor and at least one second slice tensor, the input tensor originating from the encoder path; connecting the at least one first slice tensor to the at least one second slice tensor as a function of a connection function in order to obtain a class tensor; and outputting the class tensor to the decoder path of the neural network.

Method, artificial neural network, device, computer program and machine-readable memory medium for the semantic segmentation of image data

A method for the calculation resource-saving semantic segmentation of image data of an imaging sensor with an artificial neural network, in particular, of a convolutional neural network, the artificial neural network including an encoder path, a decoder path (and a skip component), including: initial connection (merge) of an input tensor to a skip tensor with an initial connection (merge) function/connection instruction to obtain a merged tensor, the input tensor and the skip tensor being dependent on the image data; application of a function of a neural network, in particular, of a convolution to the merged tensor to obtain a proof reader tensor; second connection (merge) of the proof reader tensor to the input tensor with a second connection (merge) function/connection instruction to obtain an output tensor; outputting the output tensor to the decoder path of the artificial neural network.

A camera system is described, particularly for a vehicle, the camera system having at least: a camera which has an image sensor and a filter mask, and a control and evaluation device which receives image signals output by the image sensor, the image sensor having an array of sensor pixels outputting pixel signals, and the filter mask having an array of filters which are disposed in front of the sensor pixels and have different wavelength-selective transmission response, a portion of the filters being attenuation filters, and sensor pixels, in front of which in each case one of the attenuation filters is disposed, outputting attenuation pixel signals. The attenuation filters advantageously attenuate white light more strongly than the other filters of the filter mask; in particular, they are blue pixels.

Camera system, especially for a vehicle, and method for ascertaining image information of a signal source pulsed as a function of time

A method for detecting variable message traffic signs for a vehicle, includes: reading in a vehicle position, comparing the vehicle position to position information of at least one variable message traffic sign in order to determine a presence of a variable message traffic sign in a predefined area around the vehicle and in response to provide a proximity information, and varying a detection instruction of information of a variable message traffic sign in response to the proximity information to detect a variable message traffic sign.

Growing community of inventors

Hildesheim, Germany

Dimitrios Bariamis