arXiv subjects

Tobias Fischer

Publications and source records attributed to Tobias Fischer.

At least 19 recordsLinked to original sources

arXiv2026

Learning the Shoreline: A Very High-Resolution Approach to Reef Island Dynamics

Pacific atoll islets are often described as stable in global-scale studies, typically based on long-term shoreline proxies such as vegetation line or morphometrics like planform surface area. While informative, these approaches can obscure short-term, localized coastal dynamics -including changes in island shape and position -that are critical for ecosystem function, cultural practices, and coastal infrastructure resilience. This study presents a transferable, automated approach to shoreline monitoring using very high-resolution Pl{\'e}iades imagery and a XGBoost classifier. The method integrates spectral indices and textural features to delineate the outer limit of emerged land, including vegetated areas, beaches, man-made surfaces, and beach rock. This shoreline definition supports finescale, spatially explicit monitoring of reef island dynamics, even in morphologically complex environments. Developed and tested on multiple atolls in French Polynesia (Tetiaroa, Tikehau, Hao, and Puka Puka), the model achieves high accuracy (mean Intersection over Union $\approx$ 0.99; Mean Absolute Positional Error $\approx$ 1.28 m) and demonstrates strong performance on both training and held-out sites, validating its spatial transferability. The extracted shorelines reveal subtle but significant island-scale changes in extent, configuration, and spatial position that remain undetected by conventional shoreline proxies and surface metrics. By enabling highprecision, scalable shoreline monitoring, this method provides a more nuanced understanding of atoll change processes. It supports Pacific efforts to move beyond narratives of passive loss toward frameworks of resilience and adaptation, while providing spatial tools tailored to low-lying island realities.

eess.IV↗

arXiv2026

Why Domain Matters: Domain-Aware Benchmarking of Underwater Object Detection and Annotation Quality

Underwater object detection is strongly affected by domain shift, where performance can vary significantly across different locations, habitats, and deployment conditions. However, detector performance is typically evaluated using aggregate metrics that hide failures in specific environments, while existing domain generalization benchmarks often rely on synthetic variations that do not reflect real-world conditions. We introduce a framework that characterizes underwater images by appearance, scene composition, and acquisition geometry to assign domain labels. Using this framework, we perform the first systematic study of how domain factors influence both human annotation quality in underwater object detection datasets and deep learning-based detector performance, revealing substantial domain-dependent discrepancies. By incorporating physically meaningful domain labels, domain shift becomes something we can characterize, measure, benchmark, and act on. We highlight how this can be used to guide data collection and annotation, design more informative benchmarks, and assess detector robustness across diverse underwater environments.

cs.CV↗

arXiv2026

Strongly interacting matter with criticality induced by modified excluded volume in core-collapse supernova simulations

This article reviews critically the core-collapse supernova explosion mechanism associated with a sufficiently strong first-order phase transition from normal nuclear, in general hadronic matter to deconfined quark matter, which commonly assumes Gibbs conditions for the coexistence of phases and a phase transition construction accordingly. To this end, a novel class of multi-purpose equation of state (EOS) is developed, based on the modified excluded volume (MEV) approach employing a medium-dependent excluded-volume functional within the relativistic mean field framework with density-dependent meson-nucleon couplings. The chosen MEV parametrisation features the change in the number of degrees of freedom, mimicking the EOS softening in excess of nuclear saturation density, featuring a first-order phase transition with van der Waals like behaviour and the presence of a critical point at high temperatures. Simulations of core-collapse supernovae are performed, based on general relativistic neutrino radiation hydrodynamics in spherical symmetry, in order to explore the previously reported supernova explosion scenario within this class of phenomenological modified microscopic hadronic EOS. A burst-like neutrino signature is released, substantially longer than previously reported based on common hadron-quark hybrid model EOS with two-phase approach and Gibbs phase-transition construction, as observable signal, which is complemented by a gravitational wave mode analysis.

astro-ph.HE↗

arXiv2026

$\Lambda$ hyperons in core-collapse supernovae: Equilibration and neutrino opacities

Strange hadrons are commonly included in dense-matter equation-of-state models by imposing chemical equilibrium, but the weak-interaction timescales required to establish it in core-collapse supernovae have not been systematically assessed. In this paper we compute the $\Lambda$-hyperon production rates in the hot, dense, and isospin-asymmetric conditions characteristic of post-collapse proto-neutron stars. We find that local $\Lambda$ chemical equilibration is driven by nonleptonic strangeness-changing reactions, especially $NN\leftrightarrow N\Lambda$ scattering, on timescales of order $10^{-11}$-$10^{-10}$ s, many orders of magnitude shorter than macroscopic proto-neutron-star evolution timescales. Using an effective-field-theory framework constrained by hypernuclear weak-decay data, we find that short-range contact interactions dominate the nonleptonic rates, beyond a pure one-meson-exchange description. Semileptonic channels are too slow to set the equilibrium $\Lambda$ abundance, but they open additional absorption channels for low-energy muon neutrinos and antineutrinos, such as $\nu_\mu+\Lambda\to\mu^-+p$ and $p+\mu^-+\bar\nu_\mu\to\Lambda$. At low energies, these $\Lambda$-induced neutrino opacities exceed the corresponding nucleonic contributions for muon (anti)neutrinos, possibly influencing the evolution of the muon lepton number during proto-neutron-star deleptonization. These results support local chemical equilibrium for $\Lambda$ hyperons under the conditions studied and provide new weak-interaction input for flavor-dependent neutrino transport, muonization, and proto-neutron-star evolution.

hep-ph↗

arXiv2026

TACO: A Test and Check Framework for Robust Pose Graph Optimization

Pose Graph Optimization (PGO) is one of the most widely adopted approaches for solving Simultaneous Localization and Mapping (SLAM) problems. However, PGO approaches are particularly sensitive to outliers, which can substantially degrade the quality of the estimated trajectories. These outliers arise from incorrect place recognition associations caused by perceptual aliasing in the environment. In this paper, we present TACO (short for Test And Check Optimization), a robust optimization framework designed to filter out outliers from PGO systems. Rather than explicitly modeling measurements as inliers or outliers, TACO finds an approximation to the maximally consistent set of measurements incrementally through two complementary components: (i) The test component, namely the Incremental Probabilistic Consensus (IPC) algorithm, evaluates the consistency of each incoming loop closure online. (ii) The check component dubbed Switchable Outlier Sanitization leverages the existing Switchable Constraints to periodically sanitize any inconsistent measurements from the consistent set that IPC may have mistakenly included. We evaluate TACO on 2D SLAM and 3D Visual SLAM datasets against several state-of-the-art methods. The results show robustness comparable to state-of-the-art offline methods while preserving the computational efficiency required for online deployment, achieving a success rate above 90% in 2D and 83% in 3D across outlier rates up to 50%, with mean convergence times of approximately 45 ms and 100 ms, respectively. We release an open-source implementation of our method with this paper.

cs.RO↗

arXiv2026

DisPlace: Discriminative Place Projections for Multi-Reference Visual Place Recognition

A key challenge in Visual Place Recognition (VPR) is matching query images against reference maps captured under diverse environmental conditions and viewpoints. While multiple reference traversals improve robustness, existing fusion strategies either aggregate references uniformly or rely on heuristic selection, without distinguishing descriptor variations that preserve stable place identity from those caused by changing conditions or viewpoints. In this paper, we propose DisPlace, a multi-reference VPR framework that fuses multiple reference descriptors into a single compact and discriminative place representation. DisPlace formulates descriptor fusion as a generalized eigenvalue problem that maximizes between-place separability while suppressing within-place variation across references, rather than preserving overall descriptor variance. Unlike existing multi-reference fusion methods, DisPlace exploits variation across reference traversals to identify which linear combinations of descriptor dimensions preserve place identity and which capture condition- or viewpoint-specific variation. We evaluate DisPlace on Oxford RobotCar, Nordland, Pittsburgh30k, and Google Landmarks v2 across six state-of-the-art VPR descriptors. DisPlace outperforms seven multi-reference baselines in 49 out of 54 appearance-varying conditions, consistently improves descriptor-level fusion performance under viewpoint and unstructured settings, and requires less storage during inference than all compared fusion methods.

cs.CV↗

arXiv2026

D\'ej\`a View: Looping Transformers for Multi-View 3D Reconstruction

Recent feed-forward 3D reconstruction transformers have scaled to over a billion parameters, following the broader trend of increasing model capacity in computer vision. Yet emerging evidence suggests that contiguous transformer layers often behave like repeated applications of similar operations, and multi-view reconstruction transformers refine their predictions progressively across decoder depth. We posit that model depth partially buys iteration, paid for inefficiently in unique parameters, and instead make that iteration explicit in architecture. Our model, D\'ej\`aView, applies a single looped transformer block recurrently to per-view features for K refinement steps. Trained once, it exposes K as an inference-time compute knob, matching or outperforming substantially larger feed-forward baselines across five reconstruction benchmarks spanning indoor, outdoor, object-centric, and driving scenes, while using a fraction of their parameters and comparable or lower compute. Importantly, the same looped block formulation outperforms an otherwise identical variant with independent per-step parameters under matched training data and compute, suggesting that explicit iteration is not merely a compute-efficient substitute for capacity but a stronger inductive bias for multi-view 3D reconstruction.

cs.CV↗

arXiv2026

Charged-current neutrino opacity within the relativistic Hartree-Fock framework for astrophysical simulations of core-collapse supernovae and binary neutron star mergers

Neutrinos and their weak interactions play a vital role in the physics of core-collapse supernovae and binary neutron star mergers. Their description within astrophysical simulations, including the weak rates, is of pivotal importance not only for the prediction of accurate neutrino fluxes and spectra, including the associated conditions relevant to nucleosynthesis, neutrinos are also responsible for heating and cooling of the stellar plasma as well as the transport of lepton number and entropy. In the present article, we develop an essential improvement of the description of the underlying nuclear medium, necessary for the calculations of charged-current weak rates, with the inclusion of explicitly momentum-dependent nuclear interactions. To this end, we introduce the relativistic Hartree-Fock (RHF) approach and the associated momentum-dependent nucleon self-energies. We discuss the resulting neutrino and antineutrino opacities and find large discrepancies comparing the weak rates at the RHF level with those of commonly used relativistic mean-field (RMF) models; in particular, we observe a substantial shift of previously reported large medium-dependent modifications associated with the RMF approach.

astro-ph.HE↗

arXiv2026

Why Domain Matters: A Preliminary Study of Domain Effects in Underwater Object Detection

Domain shift, where deviations between training and deployment data distributions degrade model performance, is a key challenge in underwater environments. Existing benchmarks testing performance for underwater domain shift simulate variability through synthetic style transfer. This fails to capture intrinsic scene factors such as visibility, illumination, scene composition, or acquisition factors, limiting analysis of real-world effects. We propose a labeling framework that defines underwater domains using measurable image, scene, and acquisition characteristics. Unlike prior benchmarks, it captures physically meaningful factors, enabling semantically consistent image grouping and supporting domain-specific evaluation of detection performance including failure analysis. We validate this on public datasets, showing systematic variations across domain factors and revealing hidden failure modes.

cs.CV↗

arXiv2026

Circular polarization of gravitational waves from magnetorotational supernovae

Context. Gravitational waves (GWs) provide a unique probe of the explosion mechanism of massive stars and the evolution of nascent proto-neutron stars (PNSs). Magnetorotational explosions are one of the promising noncanonical core-collapse supernova scenarios, possibly linked to magnetar formation and energetic supernova explosions. However, the GW signatures of such events remain incompletely understood. Aims. We investigate the origin and nature of GW polarization arising from a magnetorotational core-collapse model and examine its potential detectability by current GW observatories. Methods. We performed a 3D GRMHD simulation of a rapidly rotating, strongly magnetized 20 $M_{\odot}$ progenitor, including multi-energy neutrino transport. The GW signals were extracted using the standard quadrupole formalism, and their polarization states were analyzed with Stokes parameters. Results. Strong circular polarization emerges along the rotation axis during the early post-bounce phase ($\lesssim$ 230 ms). The characteristic GW spectrum peaks at ~90 Hz, consistent with the emission at twice the local angular velocity (~45 Hz) around the PNS surface at cylindrical radii of ~50 km. These features are attributed to the low-$T/\vert{}W\vert{}$ instabilities and nonaxisymmetric motions near the PNS and not to the MHD jets themselves. The polarization signals lie within the sensitivity bands of current detectors such as Advanced LIGO, Advanced Virgo, and KAGRA. Conclusions. Models launching magnetorotationally driven jets can produce circularly polarized GW signals originating from the inner PNS region. This provides an observational signature that complements previous findings from nonmagnetized rotating models. Thus, GW polarization is a promising diagnostic of noncanonical core-collapse supernovae. Future third-generation detectors will be crucial to fully exploit this potential.

astro-ph.HE↗

arXiv2026

EventGeM: Global-to-Local Feature Matching for Event-Based Visual Place Recognition

Dynamic vision sensors, also known as event cameras, are rapidly rising in popularity for robotic and computer vision tasks due to their sparse activation and high-temporal resolution. Event cameras have been used in robotic navigation and localization tasks where accurate positioning needs to occur on small and frequent time scales, or when energy concerns are paramount. In this work, we present EventGeM, a state-of-the-art global to local feature fusion pipeline for event-based Visual Place Recognition. We use a pre-trained vision transformer (ViT-S/16) backbone to obtain global feature patch for initial match predictions embeddings from event histogram images. Local feature keypoints were then detected using a pre-trained MaxViT backbone for 2D-homography based re-ranking with RANSAC. For additional re-ranking refinement, we subsequently used a pre-trained vision foundation model for depth estimation to compare structural similarity between references and queries. Our work performs state-of-the-art localization when compared to the best currently available event-based place recognition method across several benchmark datasets and lighting conditions all whilst being fully capable of running in real-time when deployed across a variety of compute architectures. We demonstrate the capability of EventGeM in a real-world deployment on a robotic platform for online localization using event streams directly from an event camera. Project page: https://eventgemvpr.github.io/

cs.CV↗

arXiv2026

ArtiFixer: Enhancing and Extending 3D Reconstruction with Auto-Regressive Diffusion Models

Per-scene optimization methods such as 3D Gaussian Splatting provide state-of-the-art novel view synthesis quality but extrapolate poorly to under-observed areas. Methods that leverage generative priors to correct artifacts in these areas hold promise but currently suffer from two shortcomings. The first is scalability, as existing methods use image diffusion models or bidirectional video models that are limited in the number of views they can generate in a single pass (and thus require a costly iterative distillation process for consistency). The second is quality itself, as generators used in prior work tend to produce outputs that are inconsistent with existing scene content and fail entirely in completely unobserved regions. To solve these, we propose a two-stage pipeline that leverages two key insights. First, we train a powerful bidirectional generative model with a novel opacity mixing strategy that encourages consistency with existing observations while retaining the model's ability to extrapolate novel content in unseen areas. Second, we distill it into a causal auto-regressive model that generates hundreds of frames in a single pass. This model can directly produce novel views or serve as pseudo-supervision to improve the underlying 3D representation in a simple and highly efficient manner. We evaluate our method extensively and demonstrate that it can generate plausible reconstructions in scenarios where existing approaches fail completely. When measured on commonly benchmarked datasets, we outperform all existing baselines by a wide margin, exceeding prior state-of-the-art methods by 1-3 dB PSNR.

cs.CV↗

arXiv2026

Automatic Map Density Selection for Locally-Performant Visual Place Recognition

A key challenge in translating Visual Place Recognition (VPR) from the lab to long-term deployment is ensuring a priori that a system can meet user-specified performance requirements across different parts of an environment, rather than just on average globally. One critical mechanism for controlling this local performance is the density of the reference mapping database, yet this factor is largely neglected in existing work, where fixed, engineering-driven sampling densities based on sensors, storage, or GPS frequency are typically used. In this paper, we propose a VPR mapping approach that uses two reference traverses from the operating environment to automatically select an appropriate map density satisfying two user-defined requirements: (1) a target Local Recall@1 level, and (2) the proportion of the operational environment over which it must be met or exceeded, which we term the Recall Achievement Rate (RAR). Our approach uses spatial consistency and coherence features from reference-to-reference matches at multiple map densities to estimate the density needed to meet these targets on unseen deployment data. Through experiments across two VPR methods and the Nordland and Oxford RobotCar benchmarks, we show that our system consistently meets or exceeds the target Local Recall@1 over at least the user-specified proportion of the environment. Comparisons with alternative baselines show that it reliably selects an appropriate operating point in map density, avoiding unnecessarily dense maps. Finally, ablations evaluate sensitivity to reference traversal choice and segment length, and our analysis reveals that conventional global Recall@1 is a poor predictor of the often more operationally meaningful RAR metric.

cs.CV↗

arXiv2026

Long-Term Multi-Session 3D Reconstruction Under Substantial Appearance Change

Long-term environmental monitoring requires the ability to reconstruct and align 3D models across repeated site visits separated by months or years. However, existing Structure-from-Motion (SfM) pipelines implicitly assume near-simultaneous image capture and limited appearance change, and therefore fail when applied to long-term monitoring scenarios such as coral reef surveys, where substantial visual and structural change is common. In this paper, we show that the primary limitation of current approaches lies in their reliance on post-hoc alignment of independently reconstructed sessions, which is insufficient under large temporal appearance change. We address this limitation by enforcing cross-session correspondences directly within a joint SfM reconstruction. Our approach combines complementary handcrafted and learned visual features to robustly establish correspondences across large temporal gaps, enabling the reconstruction of a single coherent 3D model from imagery captured years apart, where standard independent and joint SfM pipelines break down. We evaluate our method on long-term coral reef datasets exhibiting significant real-world change, and demonstrate consistent joint reconstruction across sessions in cases where existing methods fail to produce coherent reconstructions. To ensure scalability to large datasets, we further restrict expensive learned feature matching to a small set of likely cross-session image pairs identified via visual place recognition, which reduces computational cost and improves alignment robustness.

cs.CV↗

arXiv2026

Quantile Transfer for Reliable Operating Point Selection in Visual Place Recognition

Visual Place Recognition (VPR) is a key component for localization in Global Navigation Satellite System (GNSS)-denied environments, but its performance critically depends on selecting an image matching threshold (operating point) that balances precision and recall. Thresholds are typically hand-tuned offline for a specific environment and fixed during deployment, leading to degraded performance under environmental change. We propose a method that automatically estimates the operating point of a VPR system to maximize recall whilst aiming to achieve 100% precision. The method uses a small calibration traversal with known correspondences and transfers thresholds to deployment via quantile normalization of similarity score distributions. This quantile transfer ensures that thresholds remain stable across calibration sizes and query subsets. Experiments with seven state-of-the-art VPR techniques across five benchmark datasets demonstrate that our proposed approach consistently outperforms existing baselines, enabling the underlying VPR technique to operate at 100% precision in approximately twice as many deployment scenarios (median improvement), while retrieving up to 29% more correct matches at that precision. The method eliminates manual tuning by adapting to new environments and generalizing across operating conditions. Our code is available at https://github.com/DhyeyR-007/Quantile-Transfer-for-Reliable-VPR.

cs.RO↗

arXiv2026

Ultrasensitive Real-Time Detection of SARS-CoV-2 Proteins with Arrays of Biofunctionalized Graphene Field-Effect Transistors

With the growing interest in graphene field-effect transistors (GFETs) for biosensing applications, there is a strong demand for strategies enabling flexible and multiplexed biofunctionalization, as well as highly parallel, real-time electronic readout integrated with microfluidic control. Here we present a methodology that addresses these challenges by enabling real-time, parallel monitoring of multiple GFETs integrated on a single microfabricated chip within an automated electronic and microfluidic platform. We demonstrate the capabilities of this approach through ultrasensitive detection of the SARS-CoV-2 spike (S) and nucleocapsid (N) proteins. GFET chips are functionalized via van der Waals assembly using 1 nm-thick molecular two-dimensional (2D) materials - carbon nanomembranes - which enable multiplexed biofunctionalization. The chips are integrated into a custom-developed microelectronic and microfluidic system that allows parallel, real-time, and automated measurements of 15 GFETs. We present in situ biofunctionalization of the GFETs with antibodies, followed by highly specific detection of the S- and N-proteins with limits of detection down to 10 aM and a dynamic range spanning four orders of magnitude. Owing to its versatility, the presented methodology is readily adaptable for sensing a wide range of biological and chemical targets.

physics.bio-ph↗

arXiv2025

Robust Scene Coordinate Regression via Geometrically-Consistent Global Descriptors

Recent learning-based visual localization methods use global descriptors to disambiguate visually similar places, but existing approaches often derive these descriptors from geometric cues alone (e.g., covisibility graphs), limiting their discriminative power and reducing robustness in the presence of noisy geometric constraints. We propose an aggregator module that learns global descriptors consistent with both geometrical structure and visual similarity, ensuring that images are close in descriptor space only when they are visually similar and spatially connected. This corrects erroneous associations caused by unreliable overlap scores. Using a batch-mining strategy based solely on the overlap scores and a modified contrastive loss, our method trains without manual place labels and generalizes across diverse environments. Experiments on challenging benchmarks show substantial localization gains in large-scale environments while preserving computational and memory efficiency. Code is available at https://github.com/sontung/robust_scr.

cs.CV↗

arXiv2025

ESO Expanding Horizon White Paper: Revealing the properties of matter at supranuclear densities with gravitational waves

Understanding dense matter under extreme conditions is one of the most fundamental puzzles in modern physics. Complex interactions give rise to emergent, collective phenomena. While nuclear experiments and Earth - based colliders provide valuable insights, much of the quantum chromodynamics phase diagram at high density and low temperature remains accessible only through astrophysical observations of neutron stars, neutron star mergers, and stellar collapse. Astronomical observations thus offer a direct window to the physics on subatomic scales with gravitational waves presenting an especially clean channel. Next-generation gravitational - wave observatories, such as the Einstein Telescope, would serve as unparalleled instruments to transform our understanding of neutron star matter. They will enable the detection of up to tens of thousands of binary neutron star and neutron star - black hole mergers per year, a dramatic increase over the few events accessible with current detectors. They will provide an unprecedented precision in probing cold, dense matter during the binary inspiral, exceeding by at least an order of magnitude what current facilities can achieve. Moreover, these observatories will allow us to explore uncharted regimes of dense matter at finite temperatures produced in a subset of neutron star mergers, areas that remain entirely inaccessible to current instruments. Together with multimessenger observations, these measurements will significantly deepen our knowledge of dense nuclear matter.

astro-ph.IM↗