arXiv subjects

Lars Doorenbos

Publications and source records attributed to Lars Doorenbos.

At least 19 recordsLinked to original sources

arXiv2026

Post-Training VLMs for Video Mistake Detection

Human mistakes are inevitable when following instructions, yet they can lead to severe consequences. As such, there has been an increased interest in developing methods for detecting mistakes in videos, with current methods mostly focusing on closed-set protocols. While successful in controlled settings, the closed-set assumption limits their wider applicability, as any changes to the task require collecting new data and re-training models. Instead, we argue that mistake detection methods should learn the general concept of a mistake, rather than overfitting to step-specific details. To reflect this, we introduce the Mistake Detection Video Question Answering (MD-VQA) protocol and accompanying benchmark. MD-VQA tests whether methods can discern if a step was executed correctly with respect to its description, for both seen and unseen actions. To address this important challenge, we propose the first video-language-model post-training technique for mistake detection. Our method uses a tailored reward function to encourage the model to identify discrepancies between an instruction and the corresponding video. Extensive evaluations demonstrate that this approach outperforms zero-shot, supervised fine-tuning, and post-training baselines. Notably, our method generalizes especially well to unseen procedures, for instance, with an improvement of up to 11.6% over the best-performing baseline on EP-VQA, paving the way toward general mistake detection. We release our code and benchmark at https://github.com/FedeSpu/mstk.

cs.CV↗

arXiv2026

Multi-Person Human Motion Forecasting in Complex Scenes

Accurately forecasting the movement of people in complex scenes requires reasoning over the past and present state of the entire environment. In this context, effectively incorporating object information and social interactions into a unified framework remains particularly challenging. To address this, we propose Object-Conditioned Social Diffusion (OCSD), a conditional diffusion model that integrates motion history, multi-person interactions, and object cues into a single framework. OCSD uses an object-conditioning mechanism that modulates denoising at every timestep, enabling fine-grained human-object reasoning, and a social encoder that models the interactions between all humans in the scene. As a result, our model naturally handles varying group sizes, complex social interactions, and supports sampling multiple plausible futures. Extensive experiments show that OCSD achieves state-of-the-art results on the Humans in Kitchens (HiK) and HOI-M3 benchmarks. It reduces the two-second path error by 121.5 mm (31.3%) on HiK and 130.5 mm (33.2%) on HOI-M3 compared to prior work, and produces more realistic long-term forecasts.

cs.CV↗

arXiv2026

Modality-Aware Out-of-Distribution Detection for Multi-Modal Action Recognition

The incorporation of additional modalities into action recognition models increases their performance across a wide range of settings. However, how this additional information can contribute to making the models more robust remains underexplored, particularly for the case of multi-modal out-of-distribution (OOD) detection. While methods exist that regularize the multi-modal training process with OOD detection in mind, they still apply off-the-shelf OOD detectors designed for the uni-modal case during inference, discarding important information. Based on an interesting relationship we find between the multi-modal and uni-modal predictions, we propose to use this signal to build a post-hoc detector explicitly designed for the multi-modal scenario. We combine this new source of information with a feature-space score, which detects off-manifold samples in the multi-modal space, and normalize them by the multi-modal logits. In doing so, the proposed hybrid detector is compatible with existing training-time approaches and consistently improves performance. Experiments on a wide range of established datasets from the MultiOOD benchmark show that, on average, our approach outperforms the state of the art. Our results show the importance of explicitly considering the different modalities at inference time for multi-modal OOD detection.

cs.CV↗

arXiv2026

The Unreasonable Effectiveness of VLMs for Zero-shot Procedural Mistake Detection

Procedural mistake detection is important for quality control and user assistance across many disciplines. Recent work in this field has achieved significant gains by using the reasoning capabilities of Video-Language Models (VLMs) as components within multi-stage pipelines, which consist of separate modules for supervised temporal action segmentation, error detection, and explainability. Consequently, they remain dependent on tailored training datasets and require task-specific training, limiting their wider applicability. To remedy this, we introduce zero-shot procedural mistake detection and propose a unified Zero-shot Procedural Mistake detection (ZeProM) framework that jointly solves procedural mistake detection and temporal action segmentation with a single pre-trained VLM. By evaluating our framework on two canonical mistake detection benchmarks, EgoPER and CaptainCook4D, we find that ZeProM can perform these tasks successfully, while approaching, or even outperforming, the performance of fully supervised methods. For instance, we achieve a 4.4 point improvement in EDA and a 2.0 point improvement in F1@.5 on average over all five EgoPER tasks compared to the strongest supervised methods. Overall, our results show the potential of unified methods for procedural mistake detection, and we hope this will steer the field away from highly complex pipelines and toward more generally applicable solutions.

cs.CV↗

arXiv2026

Constrained hybrid modelling to predict microbial dynamics and organic matter turnover in soil systems

Soil microorganisms control organic matter cycling and largely determine how soil systems can cope with and mitigate climate change and environmental threats. Representing microbial dynamics in process-based soil models is therefore critical to predict carbon cycling in soils, albeit highly challenging to inform from data. One promising approach to improve their parametrisation is the integration of genomic data, yet modelling the complex and unknown relationship between genomes and the processes the microbes are driving is an unsolved problem. In this work, we present the first hybrid modeling framework for deriving biokinetic parameter values of a process-based soil organic matter turnover model from metagenome-inferred functional traits based on DNA sequencing data. Our model predicts biokinetic parameters of the process-based model from genomic trait data with a neural network and integrates constraints from ecological theory and literature to ensure realistic behavior, even of non-observed state variables. We evaluate our method on synthetic genomic trait datasets of varying complexity and on real data, showing that our approach improves performance over multiple baselines and learns the dynamics of unmeasurable components of the process-based model effectively, even for small training datasets.

cs.LG↗

arXiv2026

TRAFA: Anticipating User Actions to Reduce Errors in Procedural Tasks with Predictive Feedback

Interactive assistance systems typically provide feedback after an action has been completed, supporting error recovery but not preventing the error itself. We present TRAFA, a real-time predictive feedback system for procedural tasks that intervenes before errors are committed. TRAFA operationalizes predictive feedback through a Track-Forecast-Act framework that tracks hand and object state, forecasts user motion conditioned on scene context, and triggers feedback when a predicted action is likely to violate task constraints. We instantiate this pipeline in a sequential assembly setting and evaluate it through both technical benchmarking and a controlled user study against conventional reactive feedback. Our results show that predictive feedback improves task accuracy and efficiency while maintaining a comparable number of feedback events. These findings position feedback timing as a key dimension in system design and show how real-time anticipation can be integrated into interactive systems to prevent errors before they occur.

cs.HC↗

arXiv2025

EgoControl: Controllable Egocentric Video Generation via 3D Full-Body Poses

Egocentric video generation with fine-grained control through body motion is a key requirement towards embodied AI agents that can simulate, predict, and plan actions. In this work, we propose EgoControl, a pose-controllable video diffusion model trained on egocentric data. We train a video prediction model to condition future frame generation on explicit 3D body pose sequences. To achieve precise motion control, we introduce a novel pose representation that captures both global camera dynamics and articulated body movements, and integrate it through a dedicated control mechanism within the diffusion process. Given a short sequence of observed frames and a sequence of target poses, EgoControl generates temporally coherent and visually realistic future frames that align with the provided pose control. Experimental results demonstrate that EgoControl produces high-quality, pose-consistent egocentric videos, paving the way toward controllable embodied video simulation and understanding.

cs.CV↗

arXiv2025

Video Panels for Long Video Understanding

Recent Video-Language Models (VLMs) achieve promising results on long-video understanding, but their performance still lags behind that achieved on tasks involving images or short videos. This has led to great interest in improving the long context modeling of VLMs by introducing novel modules and additional complexity. In this paper, we take a different approach: rather than fine-tuning VLMs with the limited data available, we attempt to maximize the performance of existing models. To this end, we propose a novel visual prompting strategy specifically designed for long-video understanding. By combining multiple frames as panels into one image, we effectively trade off spatial details for temporal resolution. Our approach is training-free, parameter-free, and model-agnostic, and can be seamlessly integrated into existing VLMs. Extensive experiments on five established benchmarks across a wide range of model architectures, sizes, and context windows confirm the consistency of our approach. For the TimeScope (Long) dataset, which has the longest videos, the accuracy for video question answering is improved by up to 19.4%. Overall, our method raises the bar for long video understanding models. The code is available at https://fedespu.github.io/Video-Panels.

cs.CV↗

arXiv2025

Inferring geometry and material properties from Mueller matrices with machine learning

Mueller matrices (MMs) encode information on geometry and material properties, but recovering both simultaneously is an ill-posed problem. We explore whether MMs contain sufficient information to infer surface geometry and material properties with machine learning. We use a dataset of spheres of various isotropic materials, with MMs captured over the full angular domain at five visible wavelengths (450-650 nm). We train machine learning models to predict material properties and surface normals using only these MMs as input. We demonstrate that, even when the material type is unknown, surface normals can be predicted and object geometry reconstructed. Moreover, MMs allow models to identify material types correctly. Further analyses show that diagonal elements are key for material characterization, and off-diagonal elements are decisive for normal estimation.

physics.optics↗

arXiv2025

ULISSE: Determination of star-formation rate and stellar mass based on the one-shot galaxy imaging technique

Modern sky surveys produce vast amounts of observational data, making the application of classical methods for estimating galaxy properties challenging and time-consuming. This challenge can be significantly alleviated by employing automatic machine and deep learning techniques. We propose an implementation of the ULISSE algorithm aimed at determining physical parameters of galaxies, in particular star-formation rates (SFR) and stellar masses ($M_{\ast}$), using only composite-color images. ULISSE is able to rapidly and efficiently identify candidates from a single image based on photometric and morphological similarities to a given reference object with known properties. This approach leverages features extracted from the ImageNet dataset to perform similarity searches among all objects in the sample, eliminating the need for extensive neural network training. Our experiments, performed on the Sloan Digital Sky Survey, demonstrate that we are able to predict the joint star formation rate and stellar mass of the target galaxies within 1 dex in 60% to 80% of cases, depending on the investigated subsample (quiescent/star-forming galaxies, early-/late-type, etc.), and within 0.5 dex if we consider these parameters separately. This is approximately twice the fraction obtained from a random guess extracted from the parent population. Additionally, we find ULISSE is more effective for galaxies with active star formation compared to elliptical galaxies with quenched star formation. Additionally, ULISSE performs more efficiently for galaxies with bright nuclei such as AGN. Our results suggest that ULISSE is a promising tool for a preliminary estimation of star-formation rates and stellar masses for galaxies based only on single images in current and future wide-field surveys (e.g., Euclid, LSST), which target millions of sources nightly.

astro-ph.GA↗

arXiv2024

Leveraging Transfer Learning for Astronomical Image Analysis

The exponential growth of astronomical data from large-scale surveys has created both opportunities and challenges for the astrophysics community. This paper explores the possibilities offered by transfer learning techniques in addressing these challenges across various domains of astronomical research. We present a set of recent applications of transfer learning methods for astronomical tasks based on the usage of a pre-trained convolutional neural networks. The examples shortly discussed include the detection of candidate active galactic nuclei (AGN), the possibility of deriving physical parameters for galaxies directly from images, the identification of artifacts in time series images, and the detection of strong lensing candidates and outliers. We demonstrate how transfer learning enables efficient analysis of complex astronomical phenomena, particularly in scenarios where labeled data is scarce. This kind of method will be very helpful for upcoming large-scale surveys like the Rubin Legacy Survey of Space and Time (LSST). By showcasing successful implementations and discussing methodological approaches, we highlight the versatility and effectiveness of such techniques.

astro-ph.IM↗

arXiv2024

Non-Linear Outlier Synthesis for Out-of-Distribution Detection

The reliability of supervised classifiers is severely hampered by their limitations in dealing with unexpected inputs, leading to great interest in out-of-distribution (OOD) detection. Recently, OOD detectors trained on synthetic outliers, especially those generated by large diffusion models, have shown promising results in defining robust OOD decision boundaries. Building on this progress, we present NCIS, which enhances the quality of synthetic outliers by operating directly in the diffusion's model embedding space rather than combining disjoint models as in previous work and by modeling class-conditional manifolds with a conditional volume-preserving network for more expressive characterization of the training distribution. We demonstrate that these improvements yield new state-of-the-art OOD detection results on standard ImageNet100 and CIFAR100 benchmarks and provide insights into the importance of data pre-processing and other key design choices. We make our code available at \url{https://github.com/LarsDoorenbos/NCIS}.

cs.CV↗

arXiv2024

Learning Non-Linear Invariants for Unsupervised Out-of-Distribution Detection

The inability of deep learning models to handle data drawn from unseen distributions has sparked much interest in unsupervised out-of-distribution (U-OOD) detection, as it is crucial for reliable deep learning models. Despite considerable attention, theoretically-motivated approaches are few and far between, with most methods building on top of some form of heuristic. Recently, U-OOD was formalized in the context of data invariants, allowing a clearer understanding of how to characterize U-OOD, and methods leveraging affine invariants have attained state-of-the-art results on large-scale benchmarks. Nevertheless, the restriction to affine invariants hinders the expressiveness of the approach. In this work, we broaden the affine invariants formulation to a more general case and propose a framework consisting of a normalizing flow-like architecture capable of learning non-linear invariants. Our novel approach achieves state-of-the-art results on an extensive U-OOD benchmark, and we demonstrate its further applicability to tabular data. Finally, we show our method has the same desirable properties as those based on affine invariants.

cs.CV↗

arXiv2024

Galaxy spectroscopy without spectra: Galaxy properties from photometric images with conditional diffusion models

Modern spectroscopic surveys can only target a small fraction of the vast amount of photometrically cataloged sources in wide-field surveys. Here, we report the development of a generative AI method capable of predicting optical galaxy spectra from photometric broad-band images alone. This method draws from the latest advances in diffusion models in combination with contrastive networks. We pass multi-band galaxy images into the architecture to obtain optical spectra. From these, robust values for galaxy properties can be derived with any methods in the spectroscopic toolbox, such as standard population synthesis techniques and Lick indices. When trained and tested on 64x64-pixel images from the Sloan Digital Sky Survey, the global bimodality of star-forming and quiescent galaxies in photometric space is recovered, as well as a mass-metallicity relation of star-forming galaxies. The comparison between the observed and the artificially created spectra shows good agreement in overall metallicity, age, Dn4000, stellar velocity dispersion, and E(B-V) values. Photometric redshift estimates of our generative algorithm can compete with other current, specialized deep-learning techniques. Moreover, this work is the first attempt in the literature to infer velocity dispersion from photometric images. Additionally, we can predict the presence of an active galactic nucleus up to an accuracy of 82%. With our method, scientifically interesting galaxy properties, normally requiring spectroscopic inputs, can be obtained in future data sets from large-scale photometric surveys alone. The spectra prediction via AI can further assist in creating realistic mock catalogs.

astro-ph.GA↗

arXiv2024

Iterative Deployment Exposure for Unsupervised Out-of-Distribution Detection

Deep learning models are vulnerable to performance degradation when encountering out-of-distribution (OOD) images, potentially leading to misdiagnoses and compromised patient care. These shortcomings have led to great interest in the field of OOD detection. Existing unsupervised OOD (U-OOD) detection methods typically assume that OOD samples originate from an unconcentrated distribution complementary to the training distribution, neglecting the reality that deployed models passively accumulate task-specific OOD samples over time. To better reflect this real-world scenario, we introduce Iterative Deployment Exposure (IDE), a novel and more realistic setting for U-OOD detection. We propose CSO, a method for IDE that starts from a U-OOD detector that is agnostic to the OOD distribution and slowly refines it during deployment using observed unlabeled data. CSO uses a new U-OOD scoring function that combines the Mahalanobis distance with a nearest-neighbor approach, along with a novel confidence-scaled few-shot OOD detector to effectively learn from limited OOD examples. We validate our approach on a dedicated benchmark, showing that our method greatly improves upon strong baselines on three medical imaging modalities.

cs.CV↗

arXiv2024

Identification of problematic epochs in astronomical time series through transfer learning

We present a novel method for detecting outliers in astronomical time series based on the combination of a deep neural network and a k-nearest neighbor algorithm with the aim of identifying and removing problematic epochs in the light curves of astronomical objects. We use an EfficientNet network pre-trained on ImageNet as a feature extractor and perform a k-nearest neighbor search in the resulting feature space to measure the distance from the first neighbor for each image. If the distance is above the one obtained for a stacked image, we flag the image as a potential outlier. We apply our method to time series obtained from the VLT Survey Telescope (VST) monitoring campaign of the Deep Drilling Fields of the Vera C. Rubin Legacy Survey of Space and Time (LSST). We show that our method can effectively identify and remove artifacts from the VST time series and improve the quality and reliability of the data. This approach may prove very useful in sight of the amount of data that will be provided by the LSST, which will prevent the inspection of individual light curves. We also discuss the advantages and limitations of our method and suggest possible directions for future work.

astro-ph.IM↗

arXiv2024

SuFIA: Language-Guided Augmented Dexterity for Robotic Surgical Assistants

In this work, we present SuFIA, the first framework for natural language-guided augmented dexterity for robotic surgical assistants. SuFIA incorporates the strong reasoning capabilities of large language models (LLMs) with perception modules to implement high-level planning and low-level control of a robot for surgical sub-task execution. This enables a learning-free approach to surgical augmented dexterity without any in-context examples or motion primitives. SuFIA uses a human-in-the-loop paradigm by restoring control to the surgeon in the case of insufficient information, mitigating unexpected errors for mission-critical tasks. We evaluate SuFIA on four surgical sub-tasks in a simulation environment and two sub-tasks on a physical surgical robotic platform in the lab, demonstrating its ability to perform common surgical sub-tasks through supervised autonomous operation under challenging physical and workspace conditions. Project website: orbit-surgical.github.io/sufia

cs.RO↗

arXiv2023

Stochastic Segmentation with Conditional Categorical Diffusion Models

Semantic segmentation has made significant progress in recent years thanks to deep neural networks, but the common objective of generating a single segmentation output that accurately matches the image's content may not be suitable for safety-critical domains such as medical diagnostics and autonomous driving. Instead, multiple possible correct segmentation maps may be required to reflect the true distribution of annotation maps. In this context, stochastic semantic segmentation methods must learn to predict conditional distributions of labels given the image, but this is challenging due to the typically multimodal distributions, high-dimensional output spaces, and limited annotation data. To address these challenges, we propose a conditional categorical diffusion model (CCDM) for semantic segmentation based on Denoising Diffusion Probabilistic Models. Our model is conditioned to the input image, enabling it to generate multiple segmentation label maps that account for the aleatoric uncertainty arising from divergent ground truth annotations. Our experimental results show that CCDM achieves state-of-the-art performance on LIDC, a stochastic semantic segmentation dataset, and outperforms established baselines on the classical segmentation dataset Cityscapes.

cs.CV↗