Searcharxiv⌕ Search

arXiv subjects

Stefan Becker

Publications and source records attributed to Stefan Becker.

At least 19 recordsLinked to original sources

When Does Sliced Inference Pay Off? Resolution, Tiling and Dense Fusion for Small Object Detection

Sliced inference such as SAHI is the standard remedy for small objects in high-resolution imagery, but its gains are usually measured against low-resolution baselines. We revisit sliced inference with scale as the explicit variable and ask when tiling actually pays off. We compare nine dense detectors, eight YOLO models of four generations and FCOS, on VisDrone and DOTA, under pre-specified hypotheses, calibration-only tuning and a single evaluation on held-out test data. Two findings emerge. First, much of the benefit attributed to slicing is a resolution effect: on VisDrone, a single pass trained at a higher resolution outperforms tiling with box fusion (merging decoded boxes) for every detector that could be trained at that resolution, and at equal compute a higher resolution helps more than a larger model. Second, tiles carry complementary information that box fusion wastes. We introduce SCAHI (Slicing with Coverage-Aware Hyper Inference), which accumulates the dense outputs of all passes with weights that give every image position the same total coverage, and reads objects off a peak grid fine enough to keep crowded neighbours apart. On identical passes, SCAHI outperforms box fusion for every YOLO detector on both datasets and matches it for FCOS. On DOTA, whose images far exceed the resolution a single pass can be trained at, tiles with SCAHI beat the best single pass for every detector, by 4.3 to 5.6 mAP50 at 2048 px, and replacing only SAHI's merge step by SCAHI adds 2.3 to 4.0 mAP50. An ablation attributes the gain to the border weighting and the resolution of the peak grid. Source code is available at https://github.com/beckerio/scahi.

cs.CV↗

Self-Aware Object Detection via Degradation Manifolds

Object detectors achieve strong performance under nominal imaging conditions but can fail silently when exposed to blur, noise, compression, adverse weather, or resolution changes. In safety-critical settings, it is therefore insufficient to produce predictions without assessing whether the input remains within the detector's nominal operating regime, which we refer to as self-aware object detection. We introduce a degradation-aware framework based on degradation manifolds, which structure a detector's feature space according to image degradation rather than semantic content. Our method augments a standard detection backbone with a lightweight embedding head trained via multi-layer contrastive learning, where images sharing the same degradation composition are pulled together while differing configurations are separated. Unlike prior contrastive IQA approaches, our method performs detector-aligned input-fidelity monitoring: it uses task-backbone features and a nominal prototype to detect degradation-induced shifts, rather than estimating perceptual image quality in a standalone representation. To anchor the representation, we estimate a pristine prototype from clean training embeddings, defining a nominal operating point in feature space. Self-awareness is then derived as geometric deviation from this reference, providing an intrinsic image-level signal of degradation-induced shift that is independent of detection confidence. Experiments on synthetic corruptions, cross-dataset zero-shot transfer, and natural weather-induced shifts show strong pristine-degraded separability, consistent behavior across detector architectures, and robust generalization across datasets with substantial content variation. These results suggest that degradation-aware detector representations provide a reliable signal for monitoring input-fidelity shifts.

cs.CV↗

A Causally Grounded Taxonomy for Image Degradation Robustness Evaluation

Image degradations arise throughout acquisition, processing, and transmission, altering both visual appearance and downstream machine-perception performance. They are studied across synthetic corruption benchmarks, perceptual image quality assessment (IQA), low-level vision, and physically grounded imaging-system analysis. However, these communities rely on incompatible grouping schemes and backend-specific severity definitions, making results difficult to compare across datasets, degradation models, and tasks. We propose a causally grounded framework for organizing and interpreting image degradations. Rather than introducing new degradations or redefining existing benchmarks, we provide an interpretive representation and measurement layer that makes implicit assumptions explicit. Each degradation is characterized by two orthogonal axes: its dominant causal source in the imaging pipeline and its dominant perceptual effect. This dual-axis abstraction yields a compact taxonomy spanning algorithmic corruptions, perceptual distortions, low-level imaging artifacts, and physically motivated camera or system failures. To make severity semantics observable without modifying existing implementations, we quantify each native severity level using full-reference image quality metrics: PSNR, SSIM, and LPIPS. This separates causal origin, perceptual appearance, measured distortion strength, and downstream task impact while preserving native parameterizations. We demonstrate the framework through COCO-Degradation, a taxonomy-aligned benchmark for evaluating object detector robustness under diverse low-level imaging conditions. By connecting causal source, perceptual effect, and measured severity, the proposed framework supports more interpretable and reproducible evaluation of how low-level visual degradations affect machine perception.

cs.CV↗

Disentangling coherent structures and the origin of swirl-switching

The physical origin of swirl-switching in turbulent bent-pipe flow remains the subject of ongoing research. We perform three direct numerical simulations (DNS) of flow through a $180^{\circ}$ bent pipe for curvature $γ=0.2$ at $Re_D=5,300$ and $Re_D=10,000$, and for curvature $γ=0.4$ at $Re_D=5,300$. Here, $Re_D$ is the Reynolds number based on the bulk velocity $U_b$ and pipe diameter $D$. The DNS data are subsequently used to conduct modal decompositions and local stability analysis (LSA). We discuss the limitations of common modal decomposition methods and introduce a new decomposition method, filtered Hilbert proper orthogonal decomposition (FHPOD), enabling us to isolate different instabilities as distinct FHPOD modes. FHPOD reveals seven modes belonging to four distinct families: a low-frequency axial mode, two swirl-switching modes, two swirl-breathing modes and a pair of downstream shear-layer modes. The swirl-switching and swirl-breathing modes are localised to the curved section of the bent-pipe flow and represent the sinuous and varicose instabilities of the Dean vortices, respectively. The DNS data are used to perform local stability analysis of the mean flow using the frozen eddy-viscosity method at the two streamwise wavenumbers associated with the two swirl-switching and swirl-breathing modes. The linear stability analysis reveals six unstable branches, with the swirl-switching mode being the most unstable branch. Another unstable branch represents the swirl-breathing mode. The modes obtained from the local stability analysis are in close agreement with their FHPOD counterparts in both spatial features and the range of Strouhal numbers. The results support the interpretation that swirl-switching arises as an intrinsic instability of the bent-pipe mean flow, which can be excited by incoming turbulent structures.

physics.flu-dyn↗

Comparative Study of Out-of-the-Box Technology for Automatic Target Detection and Recognition

Automatic Target Detection and Recognition (ATD/R) is critical for military decision support and (semi-)autonomous operations. Recent advances in object detection and artificial intelligence (AI) significantly boosted the potential performance of ATD/R. However, the scarcity of publicly available military datasets limits the application of these systems. As a solution, this paper explores the use of publicly available models and civilian datasets to achieve reasonable performance in military contexts. We benchmark several state-of-the-art models, including six iterations of the YOLO series and two variations on the DETR framework, on a newly acquired military relevant dataset. This dataset features military vehicles and challenging circumstances, including various degrees of occlusions and small targets. The out-of-the-box version of each model is validated alongside a version finetuned on the VisDrone dataset. This dataset features small objects, an Air-to-Ground (A2G) perspective and relevant classes, potentially generalizing to our military ATD/R task. We compare the performance of the models using mAP@0.5 and mAP@0.5:0.95, across A2G and Ground-to-Ground (G2G) perspective, target size and model size, giving insight into the real-time capabilities of models. Our main findings are: (1) bigger models outperform smaller models, (2) DETR-based models show promising results compared to the YOLO series,(3) fine-tuning models on an out-of-domain A2G dataset, improves their A2G performance and slightly improves their performance on small objects, but (4) all models still struggle with detecting small objects in an A2G scenario. We conclude that, despite recent advances in object detection, in-domain training is still crucial for creating capable ATD/R systems.

cs.CV↗

Structured Adversarial Camouflage via Voronoi Diagrams

Pixel-wise adversarial patches are computationally heavy and often visually detectable, limiting utility in security-critical systems. We present adversarial Voronoi camouflage that optimizes only seed-point locations under fixed, printable palettes using a soft assignment, producing structured, splinter camouflage-like patterns without additional regularization. Evaluated on person detection with COCO-style AP@[.5:.95], naive placement (Inria -> COCO) performs comparably bad, while garment-level application via segmentation mask (3DPeople) results in a significant AP drop. The attack transfers to out-of-domain backgrounds and across detector families (YOLOv9/10/11/12), indicating robustness in black-box settings. Repainting with different palettes largely nullifies the effect, and single-color tweaks show limited tolerance (<=0.17), highlighting a structure-palette coupling. The parameter-efficient, palette-constrained design improves visual plausibility while degrading real-time detector performance. Physical validation and color calibration are left for future work. Code: https://github.com/JensBayer/Voronoi This paper was originally presented at the International Conference on Military Communication and Information Systems (ICMCIS), organized by the Information Systems Technology (IST) Scientific and Technical Committee, IST-224-RSY - the ICMCIS, held in Bath, United Kingdom, 12-13 May 2026.

cs.CV↗

Collapse of turbulence in optimised curved pipe flow

The increased friction caused by turbulence is a significant contributor to energy consumption in the fluid-transport and piping industries. Here we describe a passive approach to reduce friction: we show that a local increase in streamwise flow curvature, combined with changing the circular cross-section to an oval, relaminarizes turbulent flow in curved pipes. We exemplify this effect in a $180^\circ$ bend at $Re_D = 10\,000$ and $20\,000$, well above the linear-stability limit. Curvature inhibits streamwise Reynolds stresses, and cross-sectional modifications weaken the secondary flow, together disrupting the near-wall regeneration cycle and collapsing turbulence. Simulations and experiments confirm that these geometric modifications suppress turbulence and reduce pressure loss by 53% and 36% compared with the baseline $180^\circ$ bend and an equal-length fully developed straight pipe, respectively. The results establish a passive, mechanism-based route to relaminarization in curved pipes with implications for energy-efficient control in other wall-bounded flows with curvature.

physics.flu-dyn↗

Higher-Order Adversarial Patches for Real-Time Object Detectors

Higher-order adversarial attacks can directly be considered the result of a cat-and-mouse game -- an elaborate action involving constant pursuit, near captures, and repeated escapes. This idiom describes the enduring circular training of adversarial attack patterns and adversarial training the best. The following work investigates the impact of higher-order adversarial attacks on object detectors by successively training attack patterns and hardening object detectors with adversarial training. The YOLOv10 object detector is chosen as a representative, and adversarial patches are used in an evasion attack manner. Our results indicate that higher-order adversarial patches are not only affecting the object detector directly trained on but rather provide a stronger generalization capacity compared to lower-order adversarial patches. Moreover, the results highlight that solely adversarial training is not sufficient to harden an object detector efficiently against this kind of adversarial attack. Code: https://github.com/JensBayer/HigherOrder

cs.CV↗

Traversing the Subspace of Adversarial Patches

Despite ongoing research on the topic of adversarial examples in deep learning for computer vision, some fundamentals of the nature of these attacks remain unclear. As the manifold hypothesis posits, high-dimensional data tends to be part of a low-dimensional manifold. To verify the thesis with adversarial patches, this paper provides an analysis of a set of adversarial patches and investigates the reconstruction abilities of three different dimensionality reduction methods. Quantitatively, the performance of reconstructed patches in an attack setting is measured and the impact of sampled patches from the latent space during adversarial training is investigated. The evaluation is performed on two publicly available datasets for person detection. The results indicate that more sophisticated dimensionality reduction methods offer no advantages over a simple principal component analysis.

cs.CV↗

Strike the Balance: On-the-Fly Uncertainty based User Interactions for Long-Term Video Object Segmentation

In this paper, we introduce a variant of video object segmentation (VOS) that bridges interactive and semi-automatic approaches, termed Lazy Video Object Segmentation (ziVOS). In contrast, to both tasks, which handle video object segmentation in an off-line manner (i.e., pre-recorded sequences), we propose through ziVOS to target online recorded sequences. Here, we strive to strike a balance between performance and robustness for long-term scenarios by soliciting user feedback's on-the-fly during the segmentation process. Hence, we aim to maximize the tracking duration of an object of interest, while requiring minimal user corrections to maintain tracking over an extended period. We propose a competitive baseline, i.e., Lazy-XMem, as a reference for future works in ziVOS. Our proposed approach uses an uncertainty estimation of the tracking state to determine whether a user interaction is necessary to refine the model's prediction. To quantitatively assess the performance of our method and the user's workload, we introduce complementary metrics alongside those already established in the field. We evaluate our approach using the recently introduced LVOS dataset, which offers numerous long-term videos. Our code is publicly available at https://github.com/Vujas-Eteph/LazyXMem.

cs.CV↗

Network transferability of adversarial patches in real-time object detection

Adversarial patches in computer vision can be used, to fool deep neural networks and manipulate their decision-making process. One of the most prominent examples of adversarial patches are evasion attacks for object detectors. By covering parts of objects of interest, these patches suppress the detections and thus make the target object 'invisible' to the object detector. Since these patches are usually optimized on a specific network with a specific train dataset, the transferability across multiple networks and datasets is not given. This paper addresses these issues and investigates the transferability across numerous object detector architectures. Our extensive evaluation across various models on two distinct datasets indicates that patches optimized with larger models provide better network transferability than patches that are optimized with smaller models.

cs.CV↗

Utilizing dataset affinity prediction in object detection to assess training data

Data pooling offers various advantages, such as increasing the sample size, improving generalization, reducing sampling bias, and addressing data sparsity and quality, but it is not straightforward and may even be counterproductive. Assessing the effectiveness of pooling datasets in a principled manner is challenging due to the difficulty in estimating the overall information content of individual datasets. Towards this end, we propose incorporating a data source prediction module into standard object detection pipelines. The module runs with minimal overhead during inference time, providing additional information about the data source assigned to individual detections. We show the benefits of the so-called dataset affinity score by automatically selecting samples from a heterogeneous pool of vehicle datasets. The results show that object detectors can be trained on a significantly sparser set of training samples without losing detection accuracy.

cs.CV↗

Generating Synthetic Ground Truth Distributions for Multi-step Trajectory Prediction using Probabilistic Composite Bézier Curves

An appropriate data basis grants one of the most important aspects for training and evaluating probabilistic trajectory prediction models based on neural networks. In this regard, a common shortcoming of current benchmark datasets is their limitation to sets of sample trajectories and a lack of actual ground truth distributions, which prevents the use of more expressive error metrics, such as the Wasserstein distance for model evaluation. Towards this end, this paper proposes a novel approach to synthetic dataset generation based on composite probabilistic Bézier curves, which is capable of generating ground truth data in terms of probability distributions over full trajectories. This allows the calculation of arbitrary posterior distributions. The paper showcases an exemplary trajectory prediction model evaluation using generated ground truth distribution data.

cs.LG↗

Vocal Fold Reconstruction from Optical Velocity and Displacement Measurements

The three-dimensional reconstruction of vocal folds in medicine usually involves endoscopy and an approach to extract depth information like structured light or stereo matching of images. The resulting mesh can accurately represent the superior area of the vocal folds, while new approaches also try to reconstruct the inferior area. We propose a novel approach to extract the time-dependent 3D geometry of the vocal fold from optical measurements on both the superior and inferior side, requiring optical measurements only from the superior side. First, a time-dependent, tri-variate surface velocity vector field is reconstructed using a high-speed camera and a laser Doppler vibrometer in an experimental environment. This vector field serves as target in an inverse finite-element simulation that optimizes the forces applied to a deformable vocal fold model such that the resulting movement after FEM simulation matches the velocity observations on the superior side. The required forces for the finite element method simulation are treated as unknowns and are assembled using multiple scalar fields. We use tensor products in Bézier Bernstein basis for our scalar fields to reduce the degrees of freedom for our optimization. We use gradient descent to optimize the control points of the force field polynomials. Our utilized error metric for gradient descent consists of two terms. The first term is used to match the simulated velocities to the observed measurements, while the second term measures the silhouette difference between observation and simulation.

physics.flu-dyn↗

READMem: Robust Embedding Association for a Diverse Memory in Unconstrained Video Object Segmentation

We present READMem (Robust Embedding Association for a Diverse Memory), a modular framework for semi-automatic video object segmentation (sVOS) methods designed to handle unconstrained videos. Contemporary sVOS works typically aggregate video frames in an ever-expanding memory, demanding high hardware resources for long-term applications. To mitigate memory requirements and prevent near object duplicates (caused by information of adjacent frames), previous methods introduce a hyper-parameter that controls the frequency of frames eligible to be stored. This parameter has to be adjusted according to concrete video properties (such as rapidity of appearance changes and video length) and does not generalize well. Instead, we integrate the embedding of a new frame into the memory only if it increases the diversity of the memory content. Furthermore, we propose a robust association of the embeddings stored in the memory with query embeddings during the update process. Our approach avoids the accumulation of redundant data, allowing us in return, to restrict the memory size and prevent extreme memory demands in long videos. We extend popular sVOS baselines with READMem, which previously showed limited performance on long videos. Our approach achieves competitive results on the Long-time Video dataset (LV1) while not hindering performance on short sequences. Our code is publicly available.

cs.CV↗

Eigenpatches -- Adversarial Patches from Principal Components

Adversarial patches are still a simple yet powerful white box attack that can be used to fool object detectors by suppressing possible detections. The patches of these so-called evasion attacks are computational expensive to produce and require full access to the attacked detector. This paper addresses the problem of computational expensiveness by analyzing 375 generated patches, calculating the principal components of these and show, that linear combinations of the resulting "eigenpatches" can be used to fool object detections successfully.

cs.CV↗

Bézier Curve Gaussian Processes

Probabilistic models for sequential data are the basis for a variety of applications concerned with processing timely ordered information. The predominant approach in this domain is given by recurrent neural networks, implementing either an approximate Bayesian approach (e.g. Variational Autoencoders or Generative Adversarial Networks) or a regression-based approach, i.e. variations of Mixture Density networks (MDN). In this paper, we focus on the $\mathcal{N}$-MDN variant, which parameterizes (mixtures of) probabilistic Bézier curves ($\mathcal{N}$-Curves) for modeling stochastic processes. While in favor in terms of computational cost and stability, MDNs generally fall behind approximate Bayesian approaches in terms of expressiveness. Towards this end, we present an approach for closing this gap by enabling full Bayesian inference on top of $\mathcal{N}$-MDNs. For this, we show that $\mathcal{N}$-Curves are a special case of Gaussian processes (denoted as $\mathcal{N}$-GP) and then derive corresponding mean and kernel functions for different modalities. Following this, we propose the use of the $\mathcal{N}$-MDN as a data-dependent generator for $\mathcal{N}$-GP prior distributions. We show the advantages granted by this combined model in an application context, using human trajectory prediction as an example.

stat.ML↗

Revisiting Click-based Interactive Video Object Segmentation

While current methods for interactive Video Object Segmentation (iVOS) rely on scribble-based interactions to generate precise object masks, we propose a Click-based interactive Video Object Segmentation (CiVOS) framework to simplify the required user workload as much as possible. CiVOS builds on de-coupled modules reflecting user interaction and mask propagation. The interaction module converts click-based interactions into an object mask, which is then inferred to the remaining frames by the propagation module. Additional user interactions allow for a refinement of the object mask. The approach is extensively evaluated on the popular interactive~DAVIS dataset, but with an inevitable adaptation of scribble-based interactions with click-based counterparts. We consider several strategies for generating clicks during our evaluation to reflect various user inputs and adjust the DAVIS performance metric to perform a hardware-independent comparison. The presented CiVOS pipeline achieves competitive results, although requiring a lower user workload.

cs.CV↗