Hamid Laga

WiFi-based 3D reconstruction aims to recover articulated human pose, dense surface correspondence, parametric meshes, depth fields, point clouds, and spatial positions or trajectories from WiFi-derived measurements, including active Channel State Information (CSI) and passive WiFi radar observations. Recent methods increasingly draw on computer vision, yet it remains unclear which principles transfer effectively and which require WiFi-specific reformulation. Existing surveys address broader wireless sensing, CSI processing, deep learning, pose recognition, reproducibility, or generalisability, but do not jointly examine multiple 3D output representations and the strength of evidence behind vision-to-WiFi transfer. This survey reviews 34 methods through two independent dimensions: reconstruction target and output representation. The taxonomy covers full-body and multi-person reconstruction, hands and local body parts, moving objects, and indoor environments, represented as 3D skeletons, dense surface correspondence, parametric meshes, depth fields, point clouds, and spatial positions or trajectories. We explicitly distinguish commodity CSI, multi-link deployments, customised antenna geometries, and passive WiFi radar systems. We also audit dataset reuse, label provenance, evaluation practices, and diagnostic controls for testing signal dependence. Four evidence-qualified lessons concern structured output priors, observation consistency and identifiability, representation--architecture alignment, and acquisition- and domain-aware modelling. Together, these findings clarify current limitations and priorities for reproducible progress.

We propose a novel framework for the statistical modeling and analysis of the spatio‐temporal shape variability in articulated 4D (i.e., 3D + time) shapes such as human bodies and animals. We treat articulated 3D shapes, represented using parametric models such as SMPL or its variants, as elements of the product space of shape and pose parameters. 4D shapes can then be seen as trajectories in this space, which has a nonlinear Riemannian structure. Our key contribution is to treat these trajectories as elements of a Riemannian shape space and propose computational tools that ( 1 ) perform temporal alignment of such trajectories to account for variations in their execution rates, ( 2 ) compute geodesics between trajectories, and thus 4D shapes, even when they exhibit different execution rates, and ( 3 ) statistically model the spatio‐temporal variability of collections of 4D shapes, enabling us to compute statistical summaries such as means and principal modes of variation. We derive a simple, yet efficient, framework for characterizing populations of 4D shapes using statistical models, which in turn can be used as a generative model for synthesizing novel 4D shapes by sampling from these distributions. We demonstrate the effectiveness of the proposed framework using publicly available 4D human and animal datasets, and show that it outperforms the state‐of‐the‐art both in terms of accuracy and computational efficiency. Our code, dataset, and videos that illustrate the results are available at https://arti4d.github.io/Arti4D/ .

As global demand for minerals in general, and critical minerals in particular, intensifies, discovering new deposits has become a pressing scientific challenge. This task requires integrating heterogeneous geoscientific data, including geophysical, geochemical, remote sensing, geological, and textual sources, across vast underexplored terrains. Such integration poses distinct challenges for Machine Learning (ML), including extreme label scarcity in a natural Positive-Unlabeled setting, spatial non-stationarity that violates independent and identically distributed (i.i.d.) assumptions, and multi-format data heterogeneity. Despite this growing body of work, a unified review connecting classical ML, deep learning, and foundation models across the full exploration pipeline is lacking. This survey reviews ML and foundation models across the mineral exploration pipeline.We cover domain-specific ML for geological, geophysical, geochemical, and remote sensing tasks; mineral prospectivity mapping (MPM) from classical methods through deep learning and 3D prediction; foundation models and NLP-based knowledge extraction across multiple geoscience modalities; and the supporting ecosystem of open datasets, portals, and software. For each area, we assess the current maturity of foundation model applications, identify domain-specific adaptations required, and highlight open gaps where AI research could have significant impact, particularly multi-modal foundation models, cross-region transfer learning, uncertainty-aware prediction, and AI agents for exploration decision support. A curated list of resources is available on the project page.

The selection of an appropriate 3D representation is a fundamental design decision that dictates the efficiency, quality, and capabilities of modern computer vision and graphics pipelines for tasks such as 3D reconstruction, novel‐view synthesis and rendering, shape and motion analysis, recognition, and generation. While traditional representations (e.g., meshes, point clouds and volumetric grids) remain standard outputs of 3D sensors (e.g., LiDAR and 3D scanners) and are widely used in downstream applications (e.g., editing and simulation), recent neural and primitive‐based representations (e.g., 3D Gaussian Splatting) offer compact and differentiable alternatives opening a wide range of opportunities in applications such as games, AR/VR, autonomous driving, robot navigation, and medical imaging, to name a few. The goal of this paper is to survey the main families of 3D representations from discrete explicit formats to continuous implicit fields based either on neural rendering or primitive splatting. For each type of representation, we present the general formulation and its variants, discuss its benefits and limitations, and highlight key applications. We conclude the paper by outlining the open challenges and potential directions for future research. Distinct from recent surveys that broadly cover 3D object and scene reconstruction, this paper provides a focused analysis on the evolution of 3D representations themselves. We specifically emphasize the paradigm shift toward implicit representations, offering a novel perspective on how these emerging formats fundamentally alter 3D/4D workflows.