arXiv subjects

Claudia Oliveira

Publications and source records attributed to Claudia Oliveira.

2 recordsLinked to original sources

arXiv2026

Polarimetry with the Giant Magellan Telescope: A Scientific Justification & Technical Requirements Document

The photons we collect from astronomical sources are encoded with information about the physics behind their origin. To fully probe that physics, we must maximize the information we extract from those photons. At optical and near-infrared wavelengths, this is typically done with the use of filters or a disperser to extract flux as a function of color. An often disregarded piece of information that is also carried by those photons is the polarization state. Astronomical polarimetry is capable of revealing details about the source geometry and magnetic fields that are often unresolvable and, therefore, inaccessible by any other means. Polarimetry is typically a photon-starved science because crucial model discriminators are often only revealed when precisions of 0.1% in the degree of polarization are achieved. For spectropolarimetry, this necessitates the collection of over one million photons at each spectral bin, which demands the use of large aperture telescopes for typical objects of interest. The future completion of Giant Segmented Mirror Telescopes (GSMTs) promises the long-awaited light grasp to initiate an unprecedented era of discovery with astronomical polarimetry.

astro-ph.IM↗

arXiv2025

CARE-PD: A Multi-Site Anonymized Clinical Dataset for Parkinson's Disease Gait Assessment

Objective gait assessment in Parkinson's Disease (PD) is limited by the absence of large, diverse, and clinically annotated motion datasets. We introduce CARE-PD, the largest publicly available archive of 3D mesh gait data for PD, and the first multi-site collection spanning 9 cohorts from 8 clinical centers. All recordings (RGB video or motion capture) are converted into anonymized SMPL meshes via a harmonized preprocessing pipeline. CARE-PD supports two key benchmarks: supervised clinical score prediction (estimating Unified Parkinson's Disease Rating Scale, UPDRS, gait scores) and unsupervised motion pretext tasks (2D-to-3D keypoint lifting and full-body 3D reconstruction). Clinical prediction is evaluated under four generalization protocols: within-dataset, cross-dataset, leave-one-dataset-out, and multi-dataset in-domain adaptation. To assess clinical relevance, we compare state-of-the-art motion encoders with a traditional gait-feature baseline, finding that encoders consistently outperform handcrafted features. Pretraining on CARE-PD reduces MPJPE (from 60.8mm to 7.5mm) and boosts PD severity macro-F1 by 17 percentage points, underscoring the value of clinically curated, diverse training data. CARE-PD and all benchmark code are released for non-commercial research at https://neurips2025.care-pd.ca/.

cs.CV↗