arXiv subjects

Maarten Bieshaar

Publications and source records attributed to Maarten Bieshaar.

At least 19 recordsLinked to original sources

arXiv2026

X-MULTI: VLM-based Imaging Factor Disentanglement for Factor-Aware Image Synthesis

Imaging factor disentanglement in text-to-image generation aims to independently control image acquisition properties such as types of camera lenses, sensor types, viewpoints, and domains to enable combinatorial generalization. This should let the model synthesize novel factor combinations unobserved in the training data, such as pairing a fisheye lens with an event sensor never observed in training data. Recent work, MULTI, introduced learnable, factor-specific embeddings to disentangle imaging factors, along with the Factor Alignment Accuracy (FAA) metric to evaluate disentanglement quality. We identify and address two independent limitations. First, MULTI's pixel-level reconstruction objective supervises the model only on observed imaging factor combinations, providing no direct training signal for novel combinations. We therefore propose X-MULTI, which uses a pretrained vision-language model (VLM) to supervise novel factor combinations synthesized during training. Second, we show the FAA metric exhibits severe cross-factor correlation leakage, misrepresenting true disentanglement quality. We therefore propose Improved-FAA (I-FAA), which employs factor-specific augmentation strategies to break these correlations and enables more rigorous evaluation. Experiments demonstrate that X-MULTI achieves improved factor alignment on novel combinations compared to MULTI. Moreover, we show that correlation leakage in FAA distorts the evaluation of true factor disentanglement and I-FAA reduces this leakage and therefore provides a more robust assessment of factor alignment.

cs.CV↗

arXiv2026

MIVIFI: Bridging Perspective and Fisheye Domains for Training Multi-View Fisheye Image Generation Models

Achieving 360{\deg} coverage is critical for the visual perception systems of autonomous vehicles. Fisheye cameras offer a cost-effective solution by enabling full surround coverage with as few as two sensors. However, existing multi-view fisheye datasets are limited, and synthesizing rare corner cases typically requires computationally expensive 3D simulations, hindering the training. While generative models have achieved significant success in standard perspective imagery, their application to wide-angle distortion remains unexplored. In this work, we formally introduce the novel problem of multi-view fisheye image generation conditioned on volumetric semantic representations and present two distinct methods. We first propose SyntheOcc-FE, which adapts the SyntheOcc architecture to fisheye data. While effective, this method is constrained by the scarcity of fisheye datasets, which limits its generalization. To overcome these limitations, we propose our second method, MIVIFI (multi-view fisheye), which leverages cross-domain learning with Equirectangular Projections. By bridging the gap between dataset domains using KITTI-360 fisheye images alongside nuScenes multi-view standard images, our approach enables high-fidelity manipulation of scene content. This framework enables the structural modification of semantic occupancy inputs to introduce or eliminate specific actors and facilitates the rendering of diverse meteorological conditions and illumination scenarios absent in the limited fisheye datasets. Quantitative and qualitative experiments demonstrate that our methods achieve robust photorealistic multi-view fisheye image generation and highlight the specific advantages of our cross-domain strategy for handling data scarcity.

cs.CV↗

arXiv2026

Self-Intersection-Aware 3D Human Motion Generation Using an Efficient Human Sphere Proxy

Human motion generation has made tremendous progress in recent years, with state-of-the-art approaches surpassing ground truth data in leading evaluation benchmarks. However, visual inspection of the generated motions paints a different picture. Even state-of-the-art approaches generate motions frequently containing self-intersections, i.e., body parts interpenetrating, which are strong artifacts, severely limiting the perceived motion quality. We introduce a novel loss, which explicitly penalizes self-intersections, to the training of human motion generation methods. We base our loss on a sphere proxy of human geometry, which allows us to calculate a self-intersection loss 98% faster and uses 83% less memory than comparable methods based on triangular meshes. The loss is agnostic to the specific approach, and we add it to the training of the recent human motion generation methods human motion diffusion model (MDM) and MoMask. Our extensive experiments show a reduction of self-intersections in generated motions of up to 49% while improving other evaluation metrics. The code is available at https://github.com/boschresearch/humansphereproxy .

cs.CV↗

arXiv2026

MULTI: Disentangling Camera Lens, Sensor, View, and Domain for Novel Image Generation

Recent text-to-image models produce high-quality images, yet text ambiguity hinders precise control when specific styles or objects are required. There have been a number of recent works dealing with learning and composing multiple objects and patterns. However, current work focuses almost entirely on image content, overlooking imaging factors such as camera lens, sensor types, imaging viewpoints, and scenes' domain characteristics. We introduce this new challenge as Imaging Factor Disentanglement and show limitations of current approaches in the regime. We, therefore, propose the new method Multi-factor disentanglement through Textual Inversion (MULTI). It consists of two stages: in the first stage, we learn general factors, and in the second stage, we extract dataset-specific ones. This setup enables the extension of existing datasets and novel factor combinations, thereby reducing distribution gaps. It further supports modifications of specific factors and image-to-image generation via ControlNets. The evaluation on our new DF-RICO benchmark demonstrates the effectiveness of MULTI and highlights the importance of Factor Disentanglement as a new direction of research.

cs.CV↗

arXiv2025

Description of Corner Cases in Automated Driving: Goals and Challenges

Scaling the distribution of automated vehicles requires handling various unexpected and possibly dangerous situations, termed corner cases (CC). Since many modules of automated driving systems are based on machine learning (ML), CC are an essential part of the data for their development. However, there is only a limited amount of CC data in large-scale data collections, which makes them challenging in the context of ML. With a better understanding of CC, offline applications, e.g., dataset analysis, and online methods, e.g., improved performance of automated driving systems, can be improved. While there are knowledge-based descriptions and taxonomies for CC, there is little research on machine-interpretable descriptions. In this extended abstract, we will give a brief overview of the challenges and goals of such a description.

cs.LG↗

arXiv2025

RICO: Two Realistic Benchmarks and an In-Depth Analysis for Incremental Learning in Object Detection

Incremental Learning (IL) trains models sequentially on new data without full retraining, offering privacy, efficiency, and scalability. IL must balance adaptability to new data with retention of old knowledge. However, evaluations often rely on synthetic, simplified benchmarks, obscuring real-world IL performance. To address this, we introduce two Realistic Incremental Object Detection Benchmarks (RICO): Domain RICO (D-RICO) features domain shifts with a fixed class set, and Expanding-Classes RICO (EC-RICO) integrates new domains and classes per IL step. Built from 14 diverse datasets covering real and synthetic domains, varying conditions (e.g., weather, time of day), camera sensors, perspectives, and labeling policies, both benchmarks capture challenges absent in existing evaluations. Our experiments show that all IL methods underperform in adaptability and retention, while replaying a small amount of previous data already outperforms all methods. However, individual training on the data remains superior. We heuristically attribute this gap to weak teachers in distillation, single models' inability to manage diverse tasks, and insufficient plasticity. Our code will be made publicly available.

cs.CV↗

arXiv2025

Incremental Object Detection with Prompt-based Methods

Visual prompt-based methods have seen growing interest in incremental learning (IL) for image classification. These approaches learn additional embedding vectors while keeping the model frozen, making them efficient to train. However, no prior work has applied such methods to incremental object detection (IOD), leaving their generalizability unclear. In this paper, we analyze three different prompt-based methods under a complex domain-incremental learning setting. We additionally provide a wide range of reference baselines for comparison. Empirically, we show that the prompt-based approaches we tested underperform in this setting. However, a strong yet practical method, combining visual prompts with replaying a small portion of previous data, achieves the best results. Together with additional experiments on prompt length and initialization, our findings offer valuable insights for advancing prompt-based IL in IOD.

cs.CV↗

arXiv2024

Criteria for Uncertainty-based Corner Cases Detection in Instance Segmentation

The operating environment of a highly automated vehicle is subject to change, e.g., weather, illumination, or the scenario containing different objects and other participants in which the highly automated vehicle has to navigate its passengers safely. These situations must be considered when developing and validating highly automated driving functions. This already poses a problem for training and evaluating deep learning models because without the costly labeling of thousands of recordings, not knowing whether the data contains relevant, interesting data for further model training, it is a guess under which conditions and situations the model performs poorly. For this purpose, we present corner case criteria based on the predictive uncertainty. With our corner case criteria, we are able to detect uncertainty-based corner cases of an object instance segmentation model without relying on ground truth (GT) data. We evaluated each corner case criterion using the COCO and the NuImages dataset to analyze the potential of our approach. We also provide a corner case decision function that allows us to distinguish each object into True Positive (TP), localization and/or classification corner case, or False Positive (FP). We also present our first results of an iterative training cycle that outperforms the baseline and where the data added to the training dataset is selected based on the corner case decision function.

cs.CV↗

arXiv2024

A Safety-Adapted Loss for Pedestrian Detection in Automated Driving

In safety-critical domains like automated driving (AD), errors by the object detector may endanger pedestrians and other vulnerable road users (VRU). As common evaluation metrics are not an adequate safety indicator, recent works employ approaches to identify safety-critical VRU and back-annotate the risk to the object detector. However, those approaches do not consider the safety factor in the deep neural network (DNN) training process. Thus, state-of-the-art DNN penalizes all misdetections equally irrespective of their criticality. Subsequently, to mitigate the occurrence of critical failure cases, i.e., false negatives, a safety-aware training strategy might be required to enhance the detection performance for critical pedestrians. In this paper, we propose a novel safety-aware loss variation that leverages the estimated per-pedestrian criticality scores during training. We exploit the reachability set-based time-to-collision (TTC-RSB) metric from the motion domain along with distance information to account for the worst-case threat quantifying the criticality. Our evaluation results using RetinaNet and FCOS on the nuScenes dataset demonstrate that training the models with our safety-aware loss function mitigates the misdetection of critical pedestrians without sacrificing performance for the general case, i.e., pedestrians outside the safety-critical zone.

cs.LG↗

arXiv2022

Space, Time, and Interaction: A Taxonomy of Corner Cases in Trajectory Datasets for Automated Driving

Trajectory data analysis is an essential component for highly automated driving. Complex models developed with these data predict other road users' movement and behavior patterns. Based on these predictions - and additional contextual information such as the course of the road, (traffic) rules, and interaction with other road users - the highly automated vehicle (HAV) must be able to reliably and safely perform the task assigned to it, e.g., moving from point A to B. Ideally, the HAV moves safely through its environment, just as we would expect a human driver to do. However, if unusual trajectories occur, so-called trajectory corner cases, a human driver can usually cope well, but an HAV can quickly get into trouble. In the definition of trajectory corner cases, which we provide in this work, we will consider the relevance of unusual trajectories with respect to the task at hand. Based on this, we will also present a taxonomy of different trajectory corner cases. The categorization of corner cases into the taxonomy will be shown with examples and is done by cause and required data sources. To illustrate the complexity between the machine learning (ML) model and the corner case cause, we present a general processing chain underlying the taxonomy.

cs.RO↗

arXiv2021

Out-of-distribution Detection and Generation using Soft Brownian Offset Sampling and Autoencoders

Deep neural networks often suffer from overconfidence which can be partly remedied by improved out-of-distribution detection. For this purpose, we propose a novel approach that allows for the generation of out-of-distribution datasets based on a given in-distribution dataset. This new dataset can then be used to improve out-of-distribution detection for the given dataset and machine learning task at hand. The samples in this dataset are with respect to the feature space close to the in-distribution dataset and therefore realistic and plausible. Hence, this dataset can also be used to safeguard neural networks, i.e., to validate the generalization performance. Our approach first generates suitable representations of an in-distribution dataset using an autoencoder and then transforms them using our novel proposed Soft Brownian Offset method. After transformation, the decoder part of the autoencoder allows for the generation of these implicit out-of-distribution samples. This newly generated dataset then allows for mixing with other datasets and thus improved training of an out-of-distribution classifier, increasing its performance. Experimentally, we show that our approach is promising for time series using synthetic data. Using our new method, we also show in a quantitative case study that we can improve the out-of-distribution detection for the MNIST dataset. Finally, we provide another case study on the synthetic generation of out-of-distribution trajectories, which can be used to validate trajectory prediction algorithms for automated driving.

cs.LG↗

arXiv2021

An Application-Driven Conceptualization of Corner Cases for Perception in Highly Automated Driving

Systems and functions that rely on machine learning (ML) are the basis of highly automated driving. An essential task of such ML models is to reliably detect and interpret unusual, new, and potentially dangerous situations. The detection of those situations, which we refer to as corner cases, is highly relevant for successfully developing, applying, and validating automotive perception functions in future vehicles where multiple sensor modalities will be used. A complication for the development of corner case detectors is the lack of consistent definitions, terms, and corner case descriptions, especially when taking into account various automotive sensors. In this work, we provide an application-driven view of corner cases in highly automated driving. To achieve this goal, we first consider existing definitions from the general outlier, novelty, anomaly, and out-of-distribution detection to show relations and differences to corner cases. Moreover, we extend an existing camera-focused systematization of corner cases by adding RADAR (radio detection and ranging) and LiDAR (light detection and ranging) sensors. For this, we describe an exemplary toolchain for data acquisition and processing, highlighting the interfaces of the corner case detection. We also define a novel level of corner cases, the method layer corner cases, which appear due to uncertainty inherent in the methodology or the data distribution.

cs.CV↗

arXiv2020

Quantile Surfaces -- Generalizing Quantile Regression to Multivariate Targets

In this article, we present a novel approach to multivariate probabilistic forecasting. Our approach is based on an extension of single-output quantile regression (QR) to multivariate-targets, called quantile surfaces (QS). QS uses a simple yet compelling idea of indexing observations of a probabilistic forecast through direction and vector length to estimate a central tendency. We extend the single-output QR technique to multivariate probabilistic targets. QS efficiently models dependencies in multivariate target variables and represents probability distributions through discrete quantile levels. Therefore, we present a novel two-stage process. In the first stage, we perform a deterministic point forecast (i.e., central tendency estimation). Subsequently, we model the prediction uncertainty using QS involving neural networks called quantile surface regression neural networks (QSNN). Additionally, we introduce new methods for efficient and straightforward evaluation of the reliability and sharpness of the issued probabilistic QS predictions. We complement this by the directional extension of the Continuous Ranked Probability Score (CRPS) score. Finally, we evaluate our novel approach on synthetic data and two currently researched real-world challenges in two different domains: First, probabilistic forecasting for renewable energy power generation, second, short-term cyclists trajectory forecasting for autonomously driving vehicles. Especially for the latter, our empirical results show that even a simple one-layer QSNN outperforms traditional parametric multivariate forecasting techniques, thus improving the state-of-the-art performance.

stat.AP↗

arXiv2020

Extended Coopetitive Soft Gating Ensemble

This article is about an extension of a recent ensemble method called Coopetitive Soft Gating Ensemble (CSGE) and its application on power forecasting as well as motion primitive forecasting of cyclists. The CSGE has been used successfully in the field of wind power forecasting, outperforming common algorithms in this domain. The principal idea of the CSGE is to weight the models regarding their observed performance during training on different aspects. Several extensions are proposed to the original CSGE within this article, making the ensemble even more flexible and powerful. The extended CSGE (XCSGE as we term it), is used to predict the power generation on both wind- and solar farms. Moreover, the XCSGE is applied to forecast the movement state of cyclists in the context of driver assistance systems. Both domains have different requirements, are non-trivial problems, and are used to evaluate various facets of the novel XCSGE. The two problems differ fundamentally in the size of the data sets and the number of features. Power forecasting is based on weather forecasts that are subject to fluctuations in their features. In the movement primitive forecasting of cyclists, time delays contribute to the difficulty of the prediction. The XCSGE reaches an improvement of the prediction performance of up to 11% for wind power forecasting and 30% for solar power forecasting compared to the worst performing model. For the classification of movement primitives of cyclists, the XCSGE reaches an improvement of up to 28%. The evaluation includes a comparison with other state-of-the-art ensemble methods. We can verify that the XCSGE results are significantly better using the Nemenyi post-hoc test.

cs.LG↗

arXiv2020

Knowledge Representations in Technical Systems -- A Taxonomy

The recent usage of technical systems in human-centric environments leads to the question, how to teach technical systems, e.g., robots, to understand, learn, and perform tasks desired by the human. Therefore, an accurate representation of knowledge is essential for the system to work as expected. This article mainly gives insight into different knowledge representation techniques and their categorization into various problem domains in artificial intelligence. Additionally, applications of presented knowledge representations are introduced in everyday robotics tasks. By means of the provided taxonomy, the search for a proper knowledge representation technique regarding a specific problem should be facilitated.

cs.AI↗

arXiv2020

Multi-Sensor Data and Knowledge Fusion -- A Proposal for a Terminology Definition

Fusion is a common tool for the analysis and utilization of available datasets and so an essential part of data mining and machine learning processes. However, a clear definition of the type of fusion is not always provided due to inconsistent literature. In the following, the process of fusion is defined depending on the fusion components and the abstraction level on which the fusion occurs. The focus in the first part of the paper at hand is on the clear definition of the terminology and the development of an appropriate ontology of the fusion components and the fusion level. In the second part, common fusion techniques are presented.

cs.AI↗

arXiv2018

Cooperative Starting Movement Detection of Cyclists Using Convolutional Neural Networks and a Boosted Stacking Ensemble

In future, vehicles and other traffic participants will be interconnected and equipped with various types of sensors, allowing for cooperation on different levels, such as situation prediction or intention detection. In this article we present a cooperative approach for starting movement detection of cyclists using a boosted stacking ensemble approach realizing feature- and decision level cooperation. We introduce a novel method based on a 3D Convolutional Neural Network (CNN) to detect starting motions on image sequences by learning spatio-temporal features. The CNN is complemented by a smart device based starting movement detection originating from smart devices carried by the cyclist. Both model outputs are combined in a stacking ensemble approach using an extreme gradient boosting classifier resulting in a fast and yet robust cooperative starting movement detector. We evaluate our cooperative approach on real-world data originating from experiments with 49 test subjects consisting of 84 starting motions.

cs.CV↗

arXiv2018

Detecting Intentions of Vulnerable Road Users Based on Collective Intelligence

Vulnerable road users (VRUs, i.e. cyclists and pedestrians) will play an important role in future traffic. To avoid accidents and achieve a highly efficient traffic flow, it is important to detect VRUs and to predict their intentions. In this article a holistic approach for detecting intentions of VRUs by cooperative methods is presented. The intention detection consists of basic movement primitive prediction, e.g. standing, moving, turning, and a forecast of the future trajectory. Vehicles equipped with sensors, data processing systems and communication abilities, referred to as intelligent vehicles, acquire and maintain a local model of their surrounding traffic environment, e.g. crossing cyclists. Heterogeneous, open sets of agents (cooperating and interacting vehicles, infrastructure, e.g. cameras and laser scanners, and VRUs equipped with smart devices and body-worn sensors) exchange information forming a multi-modal sensor system with the goal to reliably and robustly detect VRUs and their intentions under consideration of real time requirements and uncertainties. The resulting model allows to extend the perceptual horizon of the individual agent beyond their own sensory capabilities, enabling a longer forecast horizon. Concealments, implausibilities and inconsistencies are resolved by the collective intelligence of cooperating agents. Novel techniques of signal processing and modelling in combination with analytical and learning based approaches of pattern and activity recognition are used for detection, as well as intention prediction of VRUs. Cooperation, by means of probabilistic sensor and knowledge fusion, takes place on the level of perception and intention recognition. Based on the requirements of the cooperative approach for the communication a new strategy for an ad hoc network is proposed.

cs.AI↗