arXiv subjects

Joaquim Jorge

Publications and source records attributed to Joaquim Jorge.

At least 19 recordsLinked to original sources

arXiv2026

FlowSSC: Universal Generative Monocular Semantic Scene Completion via One-Step Latent Diffusion

Semantic Scene Completion (SSC) from monocular RGB images is a fundamental yet challenging task due to the inherent ambiguity of inferring occluded 3D geometry from a single view. While feed-forward methods have made progress, they often struggle to generate plausible details in occluded regions and preserve the fundamental spatial relationships of objects. Such accurate generative reasoning capability for the entire 3D space is critical in real-world applications. In this paper, we present FlowSSC, the first generative framework applied directly to monocular semantic scene completion. FlowSSC treats the SSC task as a conditional generation problem and can seamlessly integrate with existing feed-forward SSC methods to significantly boost their performance. To achieve real-time inference without compromising quality, we introduce Shortcut Flow-matching that operates in a compact triplane latent space. Unlike standard diffusion models that require hundreds of steps, our method utilizes a shortcut mechanism to achieve high-fidelity generation in a single step, enabling practical deployment in autonomous systems. Extensive experiments on SemanticKITTI demonstrate that FlowSSC achieves state-of-the-art performance, significantly outperforming existing baselines.

cs.CV↗

arXiv2025

HeadZoom: Hands-Free Zooming and Panning for 2D Image Navigation Using Head Motion

We introduce \textit{HeadZoom}, a hands-free interaction technique for navigating two-dimensional visual content using head movements. HeadZoom enables fluid zooming and panning using only real-time head tracking. It supports natural control in applications such as map exploration, radiograph inspection, and image browsing, where physical interaction is limited. We evaluated HeadZoom in a within-subjects study comparing three interaction techniques-Static, Tilt Zoom, and Parallel Zoom-across spatial, error, and subjective metrics. Parallel Zoom significantly reduced total head movement compared to Static and Tilt modes. Users reported significantly lower perceived exertion for Parallel Zoom, confirming its suitability for prolonged or precision-based tasks. By minimizing movement demands while maintaining task effectiveness, HeadZoom advances the design of head-based 2D interaction in VR and creates new opportunities for accessible hands-free systems for image exploration.

cs.HC↗

arXiv2025

SPARC: Shared Perspective with Avatar Distortion for Remote Collaboration in VR

Telepresence VR systems allow for face-to-face communication, promoting the feeling of presence and understanding of nonverbal cues. However, when discussing virtual 3D objects, limitations to presence and communication cause deictic gestures to lose meaning due to disparities in orientation. Current approaches use shared perspective, and avatar overlap to restore these references, which cause occlusions and discomfort that worsen when multiple users participate. We introduce a new approach to shared perspective in multi-user collaboration where the avatars are not co-located. Each person sees the others' avatars at their positions around the workspace while having a first-person view of the workspace. Whenever a user manipulates an object, others will see his/her arms stretching to reach that object in their perspective. SPARC combines a shared orientation and supports nonverbal communication, minimizing occlusions. We conducted a user study (n=18) to understand how the novel approach impacts task performance and workspace awareness. We found evidence that SPARC is more efficient and less mentally demanding than life-like settings.

cs.HC↗

arXiv2025

Gaze-Hand Steering for Travel and Multitasking in Virtual Environments

As head-mounted displays (HMDs) with eye-tracking become increasingly accessible, the need for effective gaze-based interfaces in virtual reality (VR) grows. Traditional gaze- or hand-based navigation often limits user precision or impairs free viewing, making multitasking difficult. We present a gaze-hand steering technique that combines eye-tracking with hand-pointing: users steer only when gaze aligns with a hand-defined target, reducing unintended actions and enabling free look. Speed is controlled via either a joystick or a waist-level speed circle. We evaluated our method in a user study (N=20) across multitasking and single-task scenarios, comparing it to a similar technique. Results show that gaze-hand steering maintains performance and enhances user comfort and spatial awareness during multitasking. Our findings support the use of gaze-hand steering in gaze-dominant VR applications requiring precision and simultaneous interaction. Our method significantly improves VR navigation in gaze-dominant, multitasking-intensive applications, supporting immersion and efficient control.

cs.HC↗

arXiv2025

DALL-M: Context-Aware Clinical Data Augmentation with LLMs

X-ray images are vital in medical diagnostics, but their effectiveness is limited without clinical context. Radiologists often find chest X-rays insufficient for diagnosing underlying diseases, necessitating the integration of structured clinical features with radiology reports. To address this, we introduce DALL-M, a novel framework that enhances clinical datasets by generating contextual synthetic data. DALL-M augments structured patient data, including vital signs (e.g., heart rate, oxygen saturation), radiology findings (e.g., lesion presence), and demographic factors. It integrates this tabular data with contextual knowledge extracted from radiology reports and domain-specific resources (e.g., Radiopaedia, Wikipedia), ensuring clinical consistency and reliability. DALL-M follows a three-phase process: (i) clinical context storage, (ii) expert query generation, and (iii) context-aware feature augmentation. Using large language models (LLMs), it generates both contextual synthetic values for existing clinical features and entirely new, clinically relevant features. Applied to 799 cases from the MIMIC-IV dataset, DALL-M expanded the original 9 clinical features to 91. Empirical validation with machine learning models (including Decision Trees, Random Forests, XGBoost, and TabNET) demonstrated a 16.5% improvement in F1 score and a 25% increase in Precision and Recall. DALL-M bridges an important gap in clinical data augmentation by preserving data integrity while enhancing predictive modeling in healthcare. Our results show that integrating LLM-generated synthetic features significantly improves model performance, making DALL-M a scalable and practical approach for AI-driven medical diagnostics.

cs.AI↗

arXiv2024

PROMPTHEUS: A Human-Centered Pipeline to Streamline SLRs with LLMs

The growing volume of academic publications poses significant challenges for researchers conducting timely and accurate Systematic Literature Reviews, particularly in fast-evolving fields like artificial intelligence. This growth of academic literature also makes it increasingly difficult for lay people to access scientific knowledge effectively, meaning academic literature is often misrepresented in the popular press and, more broadly, in society. Traditional SLR methods are labor-intensive and error-prone, and they struggle to keep up with the rapid pace of new research. To address these issues, we developed \textit{PROMPTHEUS}: an AI-driven pipeline solution that automates the SLR process using Large Language Models. We aimed to enhance efficiency by reducing the manual workload while maintaining the precision and coherence required for comprehensive literature synthesis. PROMPTHEUS automates key stages of the SLR process, including systematic search, data extraction, topic modeling using BERTopic, and summarization with transformer models. Evaluations conducted across five research domains demonstrate that PROMPTHEUS reduces review time, achieves high precision, and provides coherent topic organization, offering a scalable and effective solution for conducting literature reviews in an increasingly crowded research landscape. In addition, such tools may reduce the increasing mistrust in science by making summarization more accessible to laypeople. The code for this project can be found on the GitHub repository at https://github.com/joaopftorres/PROMPTHEUS.git

cs.AI↗

arXiv2024

4Doodle: Two-handed Gestures for Immersive Sketching of Architectural Models

Three-dimensional immersive sketching for content creation and modeling has been studied for some time. However, research in this domain mainly focused on CAVE-like scenarios. These setups can be expensive and offer a narrow interaction space. Building more affordable setups using head-mounted displays is possible, allowing greater immersion and a larger space for user physical movements. This paper presents a fully immersive environment using bi-manual gestures to sketch and create content freely in the virtual world. This approach can be applied to many scenarios, allowing people to express their ideas or review existing designs. To cope with known motor difficulties and inaccuracy of freehand 3D sketching, we explore proxy geometry and a laser-like metaphor to draw content directly from models and create content surfaces. Our current prototype offers 24 cubic meters for movement, limited by the room size. It features infinite virtual drawing space through pan and scale techniques and is larger than the typical 6-sided cave at a fraction of the cost. In a preliminary study conducted with architects and engineers, our system showed a clear promise as a tool for sketching and 3D content creation in virtual reality with a great emphasis on bi-manual gestures.

cs.HC↗

arXiv2024

Impact of Social Relationships on Peer Assessment in E-Learning

Peer assessment has been widely studied as a replacement for traditional evaluation, not only by reducing the professors' workload but mainly by benefiting students' engagement and learning. Although several works successfully validate its accuracy and fairness, more research must be done on how students' pre-existing social relationships affect the grades they give their peers in an e-learning course. We developed a Moodle plugin to provide the platform with peer assessment capabilities in forums and used it on an MSc course. The plugin curated the reviewer set for a post based on the author's relationships and included rubrics to counter the possible interpersonal effects of peer assessment. Results confirm that peer assessment is reliable and accurate for works with at least three peer assessments, although students' grades are slightly higher. The impact of social relationships is noticeable when students who do not like another peer grade their work consistently lower than students who have a positive connection. However, this has little influence on the final aggregate peer grade. Our findings show that peer assessment can replace traditional evaluation in an e-learning environment where students are familiar with each other.

cs.HC↗

arXiv2024

Benchmarking Instance-Centric Counterfactual Algorithms for XAI: From White Box to Black Box

This study investigates the impact of machine learning models on the generation of counterfactual explanations by conducting a benchmark evaluation over three different types of models: a decision tree (fully transparent, interpretable, white-box model), a random forest (semi-interpretable, grey-box model), and a neural network (fully opaque, black-box model). We tested the counterfactual generation process using four algorithms (DiCE, WatcherCF, prototype, and GrowingSpheresCF) in the literature in 25 different datasets. Our findings indicate that: (1) Different machine learning models have little impact on the generation of counterfactual explanations; (2) Counterfactual algorithms based uniquely on proximity loss functions are not actionable and will not provide meaningful explanations; (3) One cannot have meaningful evaluation results without guaranteeing plausibility in the counterfactual generation. Algorithms that do not consider plausibility in their internal mechanisms will lead to biased and unreliable conclusions if evaluated with the current state-of-the-art metrics; (4) A counterfactual inspection analysis is strongly recommended to ensure a robust examination of counterfactual explanations and the potential identification of biases.

cs.LG↗

arXiv2024

SelfReDepth: Self-Supervised Real-Time Depth Restoration for Consumer-Grade Sensors

Depth maps produced by consumer-grade sensors suffer from inaccurate measurements and missing data from either system or scene-specific sources. Data-driven denoising algorithms can mitigate such problems. However, they require vast amounts of ground truth depth data. Recent research has tackled this limitation using self-supervised learning techniques, but it requires multiple RGB-D sensors. Moreover, most existing approaches focus on denoising single isolated depth maps or specific subjects of interest, highlighting a need for methods to effectively denoise depth maps in real-time dynamic environments. This paper extends state-of-the-art approaches for depth-denoising commodity depth devices, proposing SelfReDepth, a self-supervised deep learning technique for depth restoration, via denoising and hole-filling by inpainting full-depth maps captured with RGB-D sensors. The algorithm targets depth data in video streams, utilizing multiple sequential depth frames coupled with color data to achieve high-quality depth videos with temporal coherence. Finally, SelfReDepth is designed to be compatible with various RGB-D sensors and usable in real-time scenarios as a pre-processing step before applying other depth-dependent algorithms. Our results demonstrate our approach's real-time performance on real-world datasets. They show that it outperforms state-of-the-art denoising and restoration performance at over 30fps on Commercial Depth Cameras, with potential benefits for augmented and mixed-reality applications.

cs.CV↗

arXiv2024

Pseudo-Haptics Survey: Human-Computer Interaction in Extended Reality & Teleoperation

Pseudo-haptic techniques are becoming increasingly popular in human-computer interaction. They replicate haptic sensations by leveraging primarily visual feedback rather than mechanical actuators. These techniques bridge the gap between the real and virtual worlds by exploring the brain's ability to integrate visual and haptic information. One of the many advantages of pseudo-haptic techniques is that they are cost-effective, portable, and flexible. They eliminate the need for direct attachment of haptic devices to the body, which can be heavy and large and require a lot of power and maintenance. Recent research has focused on applying these techniques to extended reality and mid-air interactions. To better understand the potential of pseudo-haptic techniques, the authors developed a novel taxonomy encompassing tactile feedback, kinesthetic feedback, and combined categories in multimodal approaches, ground not covered by previous surveys. This survey highlights multimodal strategies and potential avenues for future studies, particularly regarding integrating these techniques into extended reality and collaborative virtual environments.

cs.HC↗

arXiv2024

Eery Space: Facilitating Virtual Meetings Through Remote Proxemics

Virtual meetings have become increasingly common with modern video-conference and collaborative software. While they allow obvious savings in time and resources, current technologies add unproductive layers of protocol to the flow of communication between participants, rendering the interactions far from seamless. In this work we introduce Remote Proxemics, an extension of proxemics aimed at bringing the syntax of co-located proximal interactions to virtual meetings. We propose Eery Space, a shared virtual locus that results from merging multiple remote areas, where meeting participants' are located side-by-side as if they shared the same physical location. Eery Space promotes collaborative content creation and seamless mediation of communication channels based on virtual proximity. Results from user evaluation suggest that our approach is effective at enhancing mutual awareness between participants and sufficient to initiate proximal exchanges regardless of their geolocation, while promoting smooth interactions between local and remote people alike. These results happen even in the absence of visual avatars and other social devices such as eye contact, which are largely the focus of previous approaches.

cs.HC↗

arXiv2024

Building An Efficient Grid On GPU

Grid space partitioning is a technique to speed up queries to graphics databases. We present a parallel grid construction algorithm which can efficiently construct a structured grid on GPU hardware. Our approach is substantially faster than existing uniform grid construction algorithms, especially on non-homogeneous scenes. Indeed, it can populate a grid in real-time (at rates over 25 Hz), for architectural scenes with 10 million triangles.

cs.GR↗

arXiv2023

MDF-Net for abnormality detection by fusing X-rays with clinical data

This study investigates the effects of including patients' clinical information on the performance of deep learning (DL) classifiers for disease location in chest X-ray images. Although current classifiers achieve high performance using chest X-ray images alone, our interviews with radiologists indicate that clinical data is highly informative and essential for interpreting images and making proper diagnoses. In this work, we propose a novel architecture consisting of two fusion methods that enable the model to simultaneously process patients' clinical data (structured data) and chest X-rays (image data). Since these data modalities are in different dimensional spaces, we propose a spatial arrangement strategy, spatialization, to facilitate the multimodal learning process in a Mask R-CNN model. We performed an extensive experimental evaluation using MIMIC-Eye, a dataset comprising modalities: MIMIC-CXR (chest X-ray images), MIMIC IV-ED (patients' clinical data), and REFLACX (annotations of disease locations in chest X-rays). Results show that incorporating patients' clinical data in a DL model together with the proposed fusion methods improves the disease localization in chest X-rays by 12\% in terms of Average Precision compared to a standard Mask R-CNN using only chest X-rays. Further ablation studies also emphasize the importance of multimodal DL architectures and the incorporation of patients' clinical data in disease localization. The architecture proposed in this work is publicly available to promote the scientific reproducibility of our study (https://github.com/ChihchengHsieh/multimodal-abnormalities-detection)

eess.IV↗

arXiv2023

Development of an Immersive Virtual Colonoscopy Viewer for Colon Growths Diagnosis

Desktop-based virtual colonoscopy has been proven to be an asset in the identification of colon anomalies. The process is accurate, although time-consuming. The use of immersive interfaces for virtual colonoscopy is incipient and not yet understood. In this work, we present a new design exploring elements of the VR paradigm to make the immersive analysis more efficient while still effective. We also plan the conduction of experiments with experts to assess the multi-factor influences of coverage, duration, and diagnostic accuracy.

cs.HC↗

arXiv2023

MAGIC: Manipulating Avatars and Gestures to Improve Remote Collaboration

Remote collaborative work has become pervasive in many settings, from engineering to medical professions. Users are immersed in virtual environments and communicate through life-sized avatars that enable face-to-face collaboration. Within this context, users often collaboratively view and interact with virtual 3D models, for example, to assist in designing new devices such as customized prosthetics, vehicles, or buildings. However, discussing shared 3D content face-to-face has various challenges, such as ambiguities, occlusions, and different viewpoints that all decrease mutual awareness, leading to decreased task performance and increased errors. To address this challenge, we introduce MAGIC, a novel approach for understanding pointing gestures in a face-to-face shared 3D space, improving mutual understanding and awareness. Our approach distorts the remote userś gestures to correctly reflect them in the local userś reference space when face-to-face. We introduce a novel metric called pointing agreement to measure what two users perceive in common when using pointing gestures in a shared 3D space. Results from a user study suggest that MAGIC significantly improves pointing agreement in face-to-face collaboration settings, improving co-presence and awareness of interactions performed in the shared space. We believe that MAGIC improves remote collaboration by enabling simpler communication mechanisms and better mutual awareness.

cs.HC↗

arXiv2023

Exploring Affordances for AR in Laparoscopy

This paper explores the possibilities of designing AR interfaces to be used during laparoscopy surgery. It suggests that the laparoscopic video be displayed on AR headsets and that surgeons can consult preoperative image data on that display. Interaction with these elements is necessary, and no patterns exist to design them. Thus the paper proposes a head-gaze and clicker approach that is effective and minimalist. Finally, a prototype is presented, and an evaluation protocol is briefly discussed.

cs.HC↗

arXiv2023

SURVIVRS: Surround Video-Based Virtual Reality for Surgery Guidance

There is a strong demand for virtual reality (VR) to bring quality healthcare to underserved populations. This paper addresses this need with the design and prototype of SURVIVRS: Surround Video-Based Virtual Reality for Surgery Guidance. SURVIVRS allows a remote specialist to guide a local surgery team through a virtual reality (VR) telepresence interface. SURVIVRS is motivated by a need for medical expertise in remote and hard-to-reach areas, such as low-to-middle-income countries (LMICs). The remote surgeon interface allows the live observation of a procedure and combines 3D user interface annotation and communication tools on streams of the surgical site and the patient vitals monitor. SURVIVRS also supports debriefing and educational experiences by offering the ability for users to watch recorded surgeries from the point of view of the remote expert. The main contributions of this work are: the feasibility demonstration of the SURVIVRS system through a rigorous 3D user interface design process; the implementation of a prototype application that realizes the proposed design; and a usability evaluation of SURVIVRS showing that the tool was highly favored by users from the general population. The paper discusses the next steps in this line of research aimed at more equitable and diverse access to healthcare.

cs.HC↗