Robert Boschgmbh

Dan Zhang

Anna Khoreva

Vadim Sushko

Max Welling

Emiel Hoogeboom

Edgar Schoenfeld

Robin Tibor Schirrmeister

Christoph-Nikolas Straehle

William Harris Beluch

Kanil Patel

Jens Schwardmann

Zeynep Akata

Robert Schmier

Yuxuan Zhou

Ullrich Koethe

Nikolaj Marchenko

Jakub Tomczak

Yumeng Li

Oscar Dario Ramos Cantor

Victor Garcia Satorras

A method for training a generator for images from a semantic map that assigns each pixel of the image a semantic meaning of an object to which that pixel belongs. In the method, a mixed image is generated from an image generated by the generator and a determined actual training image, in which mixed image a first genuine subset of pixels is occupied by relevant corresponding pixel values of the image generated by the generator and the remaining genuine subset of pixels is occupied by relevant corresponding pixel values of the actual training image; and the images generated by the generator, the actual training image, and at least one mixed image, which belong to the same semantic training map, are supplied to a discriminator, which is configured to distinguish images generated by the generator from actual images of the scenery predefined by the semantic training map.

Training method for a generator for generating realistic images

A training method for training a generator neural network configured to generate synthesized sensor data. A fidelity destroying transformation is defined configured to transform a measured sensor data to obtain a fidelity-destroyed transformed measured sensor data. A fidelity preserving transformation is defined configured to transform a measured sensor data to obtain a fidelity-preserved transformed measured sensor data.

Training method for a generator neural network imposing data equivariances

A method for detecting anomalies in input image data, in particular from a camera, by detecting to what extent the input image data match at least one predefined distribution of image data or deviate from this predefined distribution. In the method: at least one transformation is provided, which maps input image data to data that have been information-reduced with regard to at least one aspect; at least one neural reconstruction network is provided, which is trained to reconstruct original image data from information-reduced data, which were obtained by applying the transformation to original image data from the predefined distribution; the input image data are mapped to information-reduced data by applying the transformation; the information-reduced data are mapped to reconstructed image data using the neural reconstruction network; the reconstructed image data are used to assess to what extent the input image data match or deviate from the predefined distribution.

Anomaly monitoring for technical systems based on the compatibility of image data with a predefined distribution

A computer-implemented method of calibrating a trained classification model. The trained classification model is trained to classify input samples according to a plurality of classes and to provide associated prediction probabilities, and includes a plurality of hidden layers and at least one activation layer. The method includes accessing the trained classification model and accessing a plurality of validation samples, each validation sample having a ground-truth label, the ground-truth label indicating a ground-truth class. The method further includes applying the trained classification model to the plurality of validation samples, obtaining, for each validation sample, an output logit vector from a layer of the trained classification model preceding a last activation layer, and training a calibration module. The calibration module is trained to adjust prediction probabilities, the prediction probabilities being derived from the output logit vectors.

Classification model calibration

A generative adversarial network. The generative adversarial network includes: a generator configured for generating an image and a corresponding label map; a discriminator configured for determining a classification of a provided image and a provided label map, wherein the classification characterizes whether the provided image and the provided label map have been generated by the generator or not and determining the classification comprises the steps of: determining a first feature map of the provided image; masking the first feature map according to the provided label map thereby determining a masked feature map; globally pooling the masked feature map thereby determining a feature representation of the provided image masked by the provided label map; determining a classification of the image based on the feature representation.

Generative adversarial network for classifying an image by masking and globally pooling a feature map

A method for training a generator for images from a semantic map that assigns to each pixel of the image a semantic meaning of an object to which this pixel belongs. The images generated by the generator and the at least one real training image that belong to the same semantic training map are supplied to a discriminator. The discriminator ascertains a semantic segmentation of the image assigned to it, the segmentation assigning a semantic meaning to each pixel of this image. From the semantic segmentation ascertained by the discriminator, it is evaluated whether the image supplied to the discriminator is a generated image or a real training image.

Training a generator for generating realistic images using a semantically segmenting discriminator

A computer-implemented method for training an augmented discriminator and a generator. The method includes: providing a training set comprising real training samples and artificial training samples for training of the augmented discriminator, wherein the artificial training samples are generated by the generator; assigning a data sequence to at least one data sample of the training set; wherein each pair of data sample and assigned data sequence is assigned to one of a plurality of classes such that, given the assigned one class and the assigned data sequence characterize whether the data sample is a real training sample or an artificial training sample; training the augmented discriminator to compute from pairs of data sample and assigned data sequence the respective class to which the corresponding pair is assigned; training the generator to generate artificial training samples such that the augmented discriminator is not able to correctly compute the aforementioned one class.

Device and method for training an augmented discriminator

A computer-implemented method for classifying whether an input signal, which comprises image and/or audio data, is anomalous or not with respect to a second data distribution using an anomaly classifier. The method includes: providing the input signal to the anomaly classifier; in the anomaly classifier, providing the input signal to a reference detector and a second detector; obtaining a reference value from the reference detector based on the input signal, the reference value characterizing the likelihood of the input signal to belong to a reference data distribution; obtaining a second value from the second detector based on the input signal, the second value characterizing the likelihood of the input signal to belong to the second data distribution; and providing an output signal, which characterizes a classification of the input signal as anomalous or not based on a comparison of the reference value and the second value.

Device and method for anomaly detection

A device and a computer implemented method of anomaly detection, including processing a digital representation of a signal or image with a wavelet decomposition to generate a first, second, and third plurality of decomposed representations, processing a first decomposed representation of the third plurality of decomposed representations with a first generative model to determine a first likelihood, processing a second decomposed representation of the third plurality of decomposed representations with a second generative model to determine a second likelihood, processing a second decomposed representation of the second plurality of decomposed representations with a third generative model to determine a third likelihood, processing a second decomposed representation of the first plurality of decomposed representations with a fourth generative model to determine a fourth likelihood, detecting an anomaly when at least one of the first likelihood, the second likelihood, the third likelihood and the fourth likelihood meets a criterium for anomaly detection.

Computer implemented method and device for anomaly detection

An apparatus () for determining a high-resolution QoS prediction map for a first radio communications network of a first environment is provided. The apparatus () comprises: a first input unit being configured to determine or provide environment information characterizing the first environment; a second input being configured to determine or provide a low-resolution QoS map associated with the first radio communications network of the first environment or a second radio communications network of a second environment; and a determination unit being configured to propagate the low-resolution QoS map and the environment information through a trained artificial deep neural network, wherein low-resolution QoS map and the environment information are provided as input parameters in an input section of the trained artificial deep neural network, and wherein the high-resolution QoS prediction map for the first radio communications network is provided in an output section of the trained artificial deep neural network.

Apparatuses and methods to determine a high-resolution QoS prediction map

A device for and a computer implemented method of digital signal processing. The method includes providing a first set of data, mapping the first set of data with to a second set of data, and determining an output of the digital signal processing depending on the second set of data. The second set of data is determined depending on a sum of a finite series of terms. At least one term of the series is determined depending on a result of a convolution of the first set of data with a kernel and at least one term of the series is determined depending on the first set of data and independent of the kernel.

Device for and computer implemented method of digital signal processing

A training system for training a generator neural network arranged to transform measured sensor data into generated sensor data. The generator network is arranged to receive as input sensor data and a transformation goal selected from a plurality of transformation goals and is arranged to transform the sensor data according to the transformation goal.

Training system for training a generator neural network

An anomaly detector () for detecting an abnormal activity in a telecommunications network. The anomaly detector () includes a receiver unit () configured to receive a plurality of network measurements (nm) in relation to the telecommunications network; a model-based mapping unit () configured to map the received plurality of network measurements (nm) via a machine-trained anomaly detection model () to an anomaly indicator (ai); and an output unit () configured to provide the anomaly indicator (ai), which indicates the abnormal activity in the telecommunications network.

Anomaly detector, anomaly detection network, method for detecting an abnormal activity, model determination unit, system, and method for determining an anomaly detection model

Computer implemented method for digital image data, digital video data or digital audio data enhancement, and a computer implemented method for encoding or decoding this data in particular for transmission or storage, wherein an element representing a part of said digital data comprises an indication of a position of the element in an ordered input data of a plurality of data elements, wherein a plurality of elements is transformed to a representation depending on an invertible linear mapping, wherein the invertible linear mapping maps the input of the plurality of elements to the representation, wherein the invertible linear mapping comprises at least one autoregressive convolution.

Growing community of inventors

Leonberg, Germany

Dan Zhang