arXiv subjects

Miguel Cazorla

Publications and source records attributed to Miguel Cazorla.

At least 19 recordsLinked to original sources

arXiv2026

VLM-as-a-Judge for Person Re-Identification

Cloth-changing person re-identification (CC-ReID) is still very challenging, and no method clearly dominates: methods that lead on one benchmark collapse on the others, because they rest on different identity cues. We show that a general-purpose vision-language model (VLM), used zero-shot as a judge for reranking a short candidate list, beats the expert retrieval models it reranks by a wide and consistent margin. On the four scenarios of CHIRLA, a long-term CC-ReID benchmark that no publicly released Re-ID model is trained on, reranking the candidates of two frozen experts raises rank-1 on every scenario, by 12.9 points on average over the best expert and 9.4 over the best naive ensemble of the two, with no component fine-tuned on the target data. Merging both experts' shortlists matches or beats reranking the stronger expert's shortlist alone, at lower cost. Introducing a VLM does, however, come at a cost: almost all of the pipeline's FLOPs are spent on it. We therefore borrow a criterion from ensembling and invoke the VLM judge only where the two experts disagree at rank 1. This threshold-free gate removes 40% of the calls, and a comparable share of total compute, with no significant drop in rank-1, trading 1.8 to 3.8 points of mAP. Judging remains the dominant cost, but on a benchmark where dedicated experts stall, we find a frozen general-purpose model to be the strongest judge of identity. The code is available at https://github.com/bdager/JudgeReID.

cs.CV↗

arXiv2026

From Frames to Temporal Graphs: In-Context Egocentric Action Recognition with Vision-Language Models

Action reasoning in egocentric video requires capturing fine-grained transitions of hand-object interactions, a task where general-purpose Vision-Language Models (VLMs) often struggle when operating directly on raw pixels. We propose to decouple visual perception from symbolic reasoning by converting videos into Temporal Action Graphs. In a multi-stage prompting pipeline, we first generate dense natural language narratives over short temporal windows as a semantic bottleneck, then formalize them into structured, open-vocabulary graph representations. On the EGTEA and Epic-Kitchens-100 datasets, the symbolic representation unlocks efficient in-context learning: few-shot graph demonstrations yield substantial accuracy gains over zero-shot frame-based inference. Even in the zero-shot setting, graph-based reasoning remains competitive with pixel-based inference despite potential pretraining contamination favoring the latter. Across 11 open-weight VLMs from 6 model families ranging from 2B to 235B parameters, our findings indicate that current VLMs are more effective as symbolic reasoners than as direct visual observers. By projecting video into the language domain, we provide a scalable, fine-tuning-free alternative to end-to-end approaches that better leverages these models' latent reasoning strengths. The code will be made public.

cs.CV↗

arXiv2026

Learning Safety-Constrained Whole-Body Loco-Manipulation for Lunar Surface Operations

Robotic systems for lunar surface operations must act autonomously and preserve their own integrity in an environment where failures are unrecoverable and continuous supervision is impossible. This paper presents a Constrained Reinforcement Learning framework for quadrupedal mobile manipulators that learns a single whole-body policy for tracking a 6D end-effector pose specified in task space, with operational limits enforced during learning through a constrained learning mechanism. We introduce a reward formulation in which the transition between locomotion and manipulation emerges from the reward geometry rather than from an explicit gating mechanism, a stability term derived from the Gravito-Inertial Inclination Margin and motivated by the reduced-gravity dynamics of the lunar surface, and a design rule for reward regularization that follows from the multiplicative structure of the termination-based objective. The framework is evaluated in a lunar-analog simulation spanning three terrain morphologies at five difficulty levels under lunar gravity, with regolith contact properties calibrated to the JSC-1a simulant and varied from loose to dense. The learned policy attains mean position and orientation errors of 3.22 cm and 6.83 deg. under nominal conditions, retains this precision across regolith densities and over the operationally representative range of terrain difficulty, and reduces soft-constraint violation time by up to an order of magnitude relative to an unconstrained baseline while incurring no hard-constraint violations.

cs.RO↗

arXiv2026

Phase-Gated Constrained Reinforcement Learning for End-to-End Capture and Detumbling of Non-Cooperative Targets

Active debris removal requires a servicer spacecraft to approach a non-cooperative tumbling object, capture it, and arrest its rotation. This sequence is conventionally divided among phase-specific controllers, each depending on knowledge of the target's inertia, a dedicated grappling fixture, or a hand-designed switch at the contact-rich seam where such pipelines are most fragile. We show that a single constrained reinforcement learning policy can execute the entire mission, driving a dual-arm free-flying servicer from free-space closing through friction-based fixture-free capture to full detumbling of a target whose mass and inertia are unknown and whose state is observed only through noisy estimates. The enabling mechanism is a smooth proximity gate applied at three loci of a constrained Markov decision process. In the reward it induces the phase structure rather than imposing it; in the action space it forecloses a degenerate grab-and-pull exploit as a hard guarantee; and in the constraint set it resolves a normalization failure of constraint-based learning, in which a constraint spanning phases that load it unequally is rendered nearly inert in the phase where precision matters most. Both the failure mode and the remedy apply to any multi-phase problem of this kind. Over 1000 randomized trials the policy succeeds in 99.5% while holding its contact-force, actuator, base-reaction and collision limits throughout, transfers zero-shot to unseen geometries and centre-of-mass offsets, and extrapolates across a tenfold range of target mass.

cs.RO↗

arXiv2026

Goal Sets, Not Goal States: Queryable Robot Goals through Goal-Set Hindsight Relabeling

Hindsight relabeling usually turns achieved future states into exact goals, which can overconstrain offline robot learning when task success depends only on a subset of the state. We propose Goal-Set Hindsight Relabeling (GS-HER), a predicate-level generalization of HER in which achieved states certify query-defined goal sets rather than singleton goal states. A binary query specifies which variables define success, making the goal predicate an inference-time input while leaving the underlying offline GCRL algorithm unchanged. Across OGBench tasks and five offline goal-conditioned learners, GS-HER improves performance when full-state goals are bottlenecked by nuisance dimensions and turns hindsight relabeling into a reusable goal interface: one checkpoint can answer multiple robot goal predicates without retraining.

cs.RO↗

arXiv2026

AIDEN: Design and Pilot Study of an AI Assistant for the Visually Impaired

This paper presents AIDEN, an artificial intelligence-based assistant designed to enhance the autonomy and daily quality of life of visually impaired individuals, who often struggle with object identification, text reading, and navigation in unfamiliar environments. Existing solutions such as screen readers or audio-based assistants facilitate access to information but frequently lead to auditory overload and raise privacy concerns in open environments. AIDEN addresses these limitations with a hybrid architecture that integrates You Only Look Once (YOLO) for real-time object detection and a Large Language and Vision Assistant (LLaVA) for scene description and Optical Character Recognition (OCR). A key novelty of the system is a continuous haptic guidance mechanism based on a Geiger-counter metaphor, which supports object centering without occupying the auditory channel, while privacy is preserved by ensuring that no personal data are stored. Empirical evaluations with visually impaired participants assessed perceived ease of use and acceptance using the Technology Acceptance Model (TAM). Results indicate high user satisfaction, particularly regarding intuitiveness and perceived autonomy. Moreover, the ``Find an Object'' achieved effective real-time performance. These findings provide promising evidence that multimodal haptic-visual feedback can improve daily usability and independence compared to traditional audio-centric methods, motivating larger-scale clinical validations.

cs.CV↗

arXiv2026

DIPSER: A Dataset for In-Person Student Engagement Recognition in the Wild

In this paper, a novel dataset is introduced, designed to assess student attention within in-person classroom settings. This dataset encompasses RGB camera data, featuring multiple cameras per student to capture both posture and facial expressions, in addition to smartwatch sensor data for each individual. This dataset allows machine learning algorithms to be trained to predict attention and correlate it with emotion. A comprehensive suite of attention and emotion labels for each student is provided, generated through self-reporting as well as evaluations by four different experts. Our dataset uniquely combines facial and environmental camera data, smartwatch metrics, and includes underrepresented ethnicities in similar datasets, all within in-the-wild, in-person settings, making it the most comprehensive dataset of its kind currently available. The dataset presented offers an extensive and diverse collection of data pertaining to student interactions across different educational contexts, augmented with additional metadata from other tools. This initiative addresses existing deficiencies by offering a valuable resource for the analysis of student attention and emotion in face-to-face lessons.

cs.CV↗

arXiv2026

Path Planning and Reinforcement Learning-Driven Control of On-Orbit Free-Flying Multi-Arm Robots

This paper presents a hybrid approach that integrates trajectory optimization (TO) and reinforcement learning (RL) for motion planning and control of free-flying multi-arm robots in on-orbit servicing scenarios. The proposed system integrates TO for generating feasible, efficient paths while accounting for dynamic and kinematic constraints, and RL for adaptive trajectory tracking under uncertainties. The multi-arm robot design, equipped with thrusters for precise body control, enables redundancy and stability in complex space operations. TO optimizes arm motions and thruster forces, reducing reliance on the arms for stabilization and enhancing maneuverability. RL further refines this by leveraging model-free control to adapt to dynamic interactions and disturbances. The experimental results validated through comprehensive simulations demonstrate the effectiveness and robustness of the proposed hybrid approach. Two case studies are explored: surface motion with initial contact and a free-floating scenario requiring surface approximation. In both cases, the hybrid method outperforms traditional strategies. In particular, the thrusters notably enhance motion smoothness, safety, and operational efficiency. The RL policy effectively tracks TO-generated trajectories, handling high-dimensional action spaces and dynamic mismatches. This integration of TO and RL combines the strengths of precise, task-specific planning with robust adaptability, ensuring high performance in the uncertain and dynamic conditions characteristic of space environments. By addressing challenges such as motion coupling, environmental disturbances, and dynamic control requirements, this framework establishes a strong foundation for advancing the autonomy and effectiveness of space robotic systems.

cs.RO↗

arXiv2026

Escaping The Big Data Paradigm in Self-Supervised Representation Learning

The reliance on large-scale datasets and extensive computational resources has become a major barrier to advancing representation learning in vision, especially in data-scarce domains. In this paper, we address the critical question: Can we escape the big data paradigm in self-supervised representation learning from images? We introduce SCOTT (Sparse Convolutional Tokenizer for Transformers), a shallow tokenization architecture that is compatible with Masked Image Modeling (MIM) tasks. SCOTT injects convolutional inductive biases into Vision Transformers (ViTs), enhancing their efficacy in small-scale data regimes. Alongside, we propose to train on a Joint-Embedding Predictive Architecture within a MIM framework (MIM-JEPA), operating in latent representation space to capture more semantic features. Our approach enables ViTs to be trained from scratch on datasets orders of magnitude smaller than traditionally required --without relying on massive external datasets for pretraining. We validate our method on three small-size, standard-resoultion, fine-grained datasets: Oxford Flowers-102, Oxford IIIT Pets-37, and ImageNet-100. Despite the challenges of limited data and high intra-class similarity, frozen SCOTT models pretrained with MIM-JEPA significantly outperform fully supervised methods and achieve competitive results with SOTA approaches that rely on large-scale pretraining, complex image augmentations and bigger model sizes. By demonstrating that robust off-the-shelf representations can be learned with limited data, compute, and model sizes, our work paves the way for computer applications in resource constrained environments such as medical imaging or robotics. Our findings challenge the prevailing notion that vast amounts of data are indispensable for effective representation learning in vision, offering a new pathway toward more accessible and inclusive advancements in the field.

cs.CV↗

arXiv2026

VLN-Pilot: Large Vision-Language Model as an Autonomous Indoor Drone Operator

This paper introduces VLN-Pilot, a novel framework in which a large Vision-and-Language Model (VLLM) assumes the role of a human pilot for indoor drone navigation. By leveraging the multimodal reasoning abilities of VLLMs, VLN-Pilot interprets free-form natural language instructions and grounds them in visual observations to plan and execute drone trajectories in GPS-denied indoor environments. Unlike traditional rule-based or geometric path-planning approaches, our framework integrates language-driven semantic understanding with visual perception, enabling context-aware, high-level flight behaviors with minimal task-specific engineering. VLN-Pilot supports fully autonomous instruction-following for drones by reasoning about spatial relationships, obstacle avoidance, and dynamic reactivity to unforeseen events. We validate our framework on a custom photorealistic indoor simulation benchmark and demonstrate the ability of the VLLM-driven agent to achieve high success rates on complex instruction-following tasks, including long-horizon navigation with multiple semantic targets. Experimental results highlight the promise of replacing remote drone pilots with a language-guided autonomous agent, opening avenues for scalable, human-friendly control of indoor UAVs in tasks such as inspection, search-and-rescue, and facility monitoring. Our results suggest that VLLM-based pilots may dramatically reduce operator workload while improving safety and mission flexibility in constrained indoor environments.

cs.RO↗

arXiv2025

Planning as Descent: Goal-Conditioned Latent Trajectory Synthesis in Learned Energy Landscapes

We present Planning as Descent (PaD), a framework for offline goal-conditioned reinforcement learning that grounds trajectory synthesis in verification. Instead of learning a policy or explicit planner, PaD learns a goal-conditioned energy function over entire latent trajectories, assigning low energy to feasible, goal-consistent futures. Planning is realized as gradient-based refinement in this energy landscape, using identical computation during training and inference to reduce train-test mismatch common in decoupled modeling pipelines. PaD is trained via self-supervised hindsight goal relabeling, shaping the energy landscape around the planning dynamics. At inference, multiple trajectory candidates are refined under different temporal hypotheses, and low-energy plans balancing feasibility and efficiency are selected. We evaluate PaD on OGBench cube manipulation tasks. When trained on narrow expert demonstrations, PaD achieves state-of-the-art 95\% success, strongly outperforming prior methods that peak at 68\%. Remarkably, training on noisy, suboptimal data further improves success and plan efficiency, highlighting the benefits of verification-driven planning. Our results suggest learning to evaluate and refine trajectories provides a robust alternative to direct policy learning for offline, reward-free planning.

cs.RO↗

arXiv2025

Simulating Students with Large Language Models: A Review of Architecture, Mechanisms, and Role Modelling in Education with Generative AI

Simulated Students offer a valuable methodological framework for evaluating pedagogical approaches and modelling diverse learner profiles, tasks which are otherwise challenging to undertake systematically in real-world settings. Recent research has increasingly focused on developing such simulated agents to capture a range of learning styles, cognitive development pathways, and social behaviours. Among contemporary simulation techniques, the integration of large language models (LLMs) into educational research has emerged as a particularly versatile and scalable paradigm. LLMs afford a high degree of linguistic realism and behavioural adaptability, enabling agents to approximate cognitive processes and engage in contextually appropriate pedagogical dialogues. This paper presents a thematic review of empirical and methodological studies utilising LLMs to simulate student behaviour across educational environments. We synthesise current evidence on the capacity of LLM-based agents to emulate learner archetypes, respond to instructional inputs, and interact within multi-agent classroom scenarios. Furthermore, we examine the implications of such systems for curriculum development, instructional evaluation, and teacher training. While LLMs surpass rule-based systems in natural language generation and situational flexibility, ongoing concerns persist regarding algorithmic bias, evaluation reliability, and alignment with educational objectives. The review identifies existing technological and methodological gaps and proposes future research directions for integrating generative AI into adaptive learning systems and instructional design.

cs.CY↗

arXiv2025

Autonomous Legged Mobile Manipulation for Lunar Surface Operations via Constrained Reinforcement Learning

Robotics plays a pivotal role in planetary science and exploration, where autonomous and reliable systems are crucial due to the risks and challenges inherent to space environments. The establishment of permanent lunar bases demands robotic platforms capable of navigating and manipulating in the harsh lunar terrain. While wheeled rovers have been the mainstay for planetary exploration, their limitations in unstructured and steep terrains motivate the adoption of legged robots, which offer superior mobility and adaptability. This paper introduces a constrained reinforcement learning framework designed for autonomous quadrupedal mobile manipulators operating in lunar environments. The proposed framework integrates whole-body locomotion and manipulation capabilities while explicitly addressing critical safety constraints, including collision avoidance, dynamic stability, and power efficiency, in order to ensure robust performance under lunar-specific conditions, such as reduced gravity and irregular terrain. Experimental results demonstrate the framework's effectiveness in achieving precise 6D task-space end-effector pose tracking, achieving an average positional accuracy of 4 cm and orientation accuracy of 8.1 degrees. The system consistently respects both soft and hard constraints, exhibiting adaptive behaviors optimized for lunar gravity conditions. This work effectively bridges adaptive learning with essential mission-critical safety requirements, paving the way for advanced autonomous robotic explorers for future lunar missions.

cs.RO↗

arXiv2025

CHIRLA: Comprehensive High-resolution Identification and Re-identification for Large-scale Analysis

Person re-identification (Re-ID) is a key challenge in computer vision, requiring the matching of individuals across cameras, locations, and time. While most research focuses on short-term scenarios with minimal appearance changes, real-world applications demand robust systems that handle long-term variations caused by clothing and physical changes. We present CHIRLA, Comprehensive High-resolution Identification and Re-identification for Large-scale Analysis, a novel dataset designed for video-based long-term person Re-ID. CHIRLA was recorded over seven months in four connected indoor environments using seven strategically placed cameras, capturing realistic movements with substantial clothing and appearance variability. The dataset includes 22 individuals, more than five hours of video, and about 1M bounding boxes with identity annotations obtained through semi-automatic labeling. We also define benchmark protocols for person tracking and Re-ID, covering diverse and challenging scenarios such as occlusion, reappearance, and multi-camera conditions. By introducing this comprehensive benchmark, we aim to facilitate the development and evaluation of Re-ID algorithms that can reliably perform in challenging, long-term real-world scenarios. The benchmark code is publicly available at: https://github.com/bdager/CHIRLA.

cs.CV↗

arXiv2025

UKDM: Underwater keypoint detection and matching using underwater image enhancement techniques

The purpose of this paper is to explore the use of underwater image enhancement techniques to improve keypoint detection and matching. By applying advanced deep learning models, including generative adversarial networks and convolutional neural networks, we aim to find the best method which improves the accuracy of keypoint detection and the robustness of matching algorithms. We evaluate the performance of these techniques on various underwater datasets, demonstrating significant improvements over traditional methods.

cs.CV↗

arXiv2025

CADDI: An in-Class Activity Detection Dataset using IMU data from low-cost sensors

The monitoring and prediction of in-class student activities is of paramount importance for the comprehension of engagement and the enhancement of pedagogical efficacy. The accurate detection of these activities enables educators to modify their lessons in real time, thereby reducing negative emotional states and enhancing the overall learning experience. To this end, the use of non-intrusive devices, such as inertial measurement units (IMUs) embedded in smartwatches, represents a viable solution. The development of reliable predictive systems has been limited by the lack of large, labeled datasets in education. To bridge this gap, we present a novel dataset for in-class activity detection using affordable IMU sensors. The dataset comprises 19 diverse activities, both instantaneous and continuous, performed by 12 participants in typical classroom scenarios. It includes accelerometer, gyroscope, rotation vector data, and synchronized stereo images, offering a comprehensive resource for developing multimodal algorithms using sensor and visual data. This dataset represents a key step toward scalable solutions for activity recognition in educational settings.

cs.CV↗

arXiv2021

NVS-MonoDepth: Improving Monocular Depth Prediction with Novel View Synthesis

Building upon the recent progress in novel view synthesis, we propose its application to improve monocular depth estimation. In particular, we propose a novel training method split in three main steps. First, the prediction results of a monocular depth network are warped to an additional view point. Second, we apply an additional image synthesis network, which corrects and improves the quality of the warped RGB image. The output of this network is required to look as similar as possible to the ground-truth view by minimizing the pixel-wise RGB reconstruction error. Third, we reapply the same monocular depth estimation onto the synthesized second view point and ensure that the depth predictions are consistent with the associated ground truth depth. Experimental results prove that our method achieves state-of-the-art or comparable performance on the KITTI and NYU-Depth-v2 datasets with a lightweight and simple vanilla U-Net architecture.

cs.CV↗

arXiv2021

A Large Visual, Qualitative and Quantitative Dataset of Web Pages

The World Wide Web is not only one of the most important platforms of communication and information at present, but also an area of growing interest for scientific research. This motivates a lot of work and projects that require large amounts of data. However, there is no dataset that integrates the parameters and visual appearance of Web pages, because its collection is a costly task in terms of time and effort. With the support of various computer tools and programming scripts, we have created a large dataset of 49,438 Web pages. It consists of visual, textual and numerical data types, includes all countries worldwide, and considers a broad range of topics such as art, entertainment, economy, business, education, government, news, media, science, and environment, covering different cultural characteristics and varied design preferences. In this paper, we describe the process of collecting, debugging and publishing the final product, which is freely available. To demonstrate the usefulness of our dataset, we expose a binary classification model for detecting error Web pages, and a multi-class Web subject-based categorization, both problems using convolutional neural networks.

cs.CV↗