Abstract: A method for training a machine learning (ML) model for generating a joint particle-based representation of a scene from multi-modality sensor data. The method includes receiving sensor data acquired by a set of optical sensors of different sensor modality, representing the same scene. A joint particle-based representation of the scene covered by the optical sensors is generated. Intersections between rays emanating from a sensor location with the joint particle-based representation are determined, based on which an observation of the scene is rendered. A sensor-specific loss function value is determined by comparing the rendered observation with an observation rendered based on the received sensor data. The ML model is trained by determining the intersections, rendering the observation and determining the value of the sensor-specific loss function for every optical sensor, and by optimizing a combination of the sensor-specific loss functions.

Abstract: A method for training a neural network that predicts one or more characteristic values of a specified scenery.

Abstract: A method for ascertaining an error measure in the detection and/or classification of image data in a predetermined domain. The method includes: providing image data for the predetermined domain; applying an image perturbation to the provided image data to generate perturbed image data; providing the perturbed image data to a pre-trained perturbation removal diffusion model; removing the perturbations from the perturbed image data by the pre-trained perturbation removal diffusion model in order to generate perturbation-removed image data; providing the perturbation-removed image data to a detection and/or classification model for detecting the domain in the perturbation-removed image data and/or for classifying the perturbation-removed image data into a class of the domain; and ascertaining an error measure of the detection and/or the classification using the detected domain or the classified class of the domain and the image perturbations.

Abstract: Method and apparatus for improving synthetic ground truth data by means of a data generator and for training a target machine learning model. The method includes: providing ground truth data samples which relate to ground truth source data and are synthetically generated by the data generator; comparing a performance of the data generator for the provided ground truth data samples with a performance threshold value; generating anew ground truth data samples for the same ground truth source data by means of the data generator if the performance threshold value for the provided ground truth data samples is not achieved; replacing the ground truth data samples for which the performance threshold value is not achieved with the newly generated ground truth data samples; and training the target machine learning model on the basis of the replaced and provided ground truth data samples.