arXiv subjects

Francisco Gomez-Donoso

Publications and source records attributed to Francisco Gomez-Donoso.

7 recordsLinked to original sources

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 and graph-based inference alike. 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

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

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

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

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↗

arXiv2020

Attentional-GCNN: Adaptive Pedestrian Trajectory Prediction towards Generic Autonomous Vehicle Use Cases

Autonomous vehicle navigation in shared pedestrian environments requires the ability to predict future crowd motion both accurately and with minimal delay. Understanding the uncertainty of the prediction is also crucial. Most existing approaches however can only estimate uncertainty through repeated sampling of generative models. Additionally, most current predictive models are trained on datasets that assume complete observability of the crowd using an aerial view. These are generally not representative of real-world usage from a vehicle perspective, and can lead to the underestimation of uncertainty bounds when the on-board sensors are occluded. Inspired by prior work in motion prediction using spatio-temporal graphs, we propose a novel Graph Convolutional Neural Network (GCNN)-based approach, Attentional-GCNN, which aggregates information of implicit interaction between pedestrians in a crowd by assigning attention weight in edges of the graph. Our model can be trained to either output a probabilistic distribution or faster deterministic prediction, demonstrating applicability to autonomous vehicle use cases where either speed or accuracy with uncertainty bounds are required. To further improve the training of predictive models, we propose an automatically labelled pedestrian dataset collected from an intelligent vehicle platform representative of real-world use. Through experiments on a number of datasets, we show our proposed method achieves an improvement over the state of art by 10% Average Displacement Error (ADE) and 12% Final Displacement Error (FDE) with fast inference speeds.

cs.CV↗

arXiv2017

Large-scale Multiview 3D Hand Pose Dataset

Accurate hand pose estimation at joint level has several uses on human-robot interaction, user interfacing and virtual reality applications. Yet, it currently is not a solved problem. The novel deep learning techniques could make a great improvement on this matter but they need a huge amount of annotated data. The hand pose datasets released so far present some issues that make them impossible to use on deep learning methods such as the few number of samples, high-level abstraction annotations or samples consisting in depth maps. In this work, we introduce a multiview hand pose dataset in which we provide color images of hands and different kind of annotations for each, i.e the bounding box and the 2D and 3D location on the joints in the hand. Besides, we introduce a simple yet accurate deep learning architecture for real-time robust 2D hand pose estimation.

cs.HC↗