arXiv subjects

Mattia Rossi

Publications and source records attributed to Mattia Rossi.

At least 19 recordsLinked to original sources

arXiv2026

The recent crossing of the 7:3 resonance between Ganymede and Callisto

Io, Europa and Ganymede are locked in the well-known Laplace resonance, while Callisto is not, even though it is very close to the 7:3 mean motion resonance with Ganymede. When considering the estimated tidal dissipation in the Jovian system and the consequent orbital migration rate of the moons, it follows that the 7:3 resonance should have been encountered around two million years ago. Through accurate numerical simulations, we investigated the dynamical effects of the 7:3 resonance crossing on the recent orbital evolution of the Galilean moons. We found that the most probable dynamical pathway is that Ganymede and Callisto were not captured into resonance, not even temporarily, while their eccentricities experienced a downward kick because of the resonance crossing, reducing their mean values of about 16% and 5%, respectively. Since this kind of evolution requires a value of the eccentricity of Ganymede before the resonant encounter slightly larger than today's, it follows that tidal dissipation within the moon must be low. In fact, a null free eccentricity of Ganymede would have led to an almost certain capture into the 7:3 resonance that would have survived until today. As a result of the resonance encounter, the amplitude of the free libration of the Laplace resonant angle increased, providing an explanation for its current value. Moreover, we found evidence that a three-body resonance crossing between the three outer moons occurred within the last few tens of thousands of years.

astro-ph.EP↗

arXiv2026

Learning Spectral and Polarimetric Clues for One-to-Multimodal Novel View Synthesis

Neural rendering techniques allow for accurate reconstruction of the geometry and color appearance of 3D scenes. Some methods have extended their use to additional imaging modalities, such as multispectral, infrared, or polarimetric data. However, all of these approaches require expensive sensors and calibrated setups to capture new multimodal frames for each new scene. We propose Spectral and Polarimetric Implicit Learned Representation (SPoILeR), a novel method to obtain multi-view consistent renderings of unconventional modalities for scenes where either only RGB frames or very few of the additional modalities are available. Thanks to a multimodal pre-training phase, the model learns the mutual correlation between different modalities. This step allows predicting accurate renderings of unconventional modalities during a fine-tuning phase supervised only by RGB images. Experimental results show that the approach can accurately render infrared, polarimetric, and multispectral frames for scenes where no input sample captured by these types of sensors is provided.

cs.CV↗

arXiv2026

EPO: Boosting 3D Foundation Models with Edge-based Pose Optimization

We introduce \textbf{Edge-based Pose Optimization (EPO)}, a trackless geometric optimization framework specifically designed to boost the Structure-from-Motion reconstructions generated by 3D Foundation Models. These models achieve rapid inference by bypassing the time-consuming feature extraction and matching stages of traditional pipelines, where explicit correspondences between each 3D point and multiple images, referred to as tracks, are established. However, their geometric accuracy currently falls short of traditional pipelines. While this can be addressed in a post-processing step via Bundle Adjustment-like refinement, doing so requires extracting feature tracks, thus defeating the original speed advantage. Instead, our fully differentiable framework uses edge map alignment as a proxy for geometric optimization, avoiding feature extraction and track construction entirely. Through extensive evaluation across multiple datasets and tasks, we demonstrate that EPO matches or outperforms Bundle Adjustment-like methods while requiring significantly lower runtime and memory. Notably, its reduced memory footprint makes EPO suitable for consumer-grade hardware, where competing refinement methods cannot run.

cs.CV↗

arXiv2026

TerraSky3D: Multi-View Reconstructions of European Landmarks in 4K

Despite the growing need for data of more and more sophisticated 3D reconstruction pipelines, we can still observe a scarcity of suitable public datasets. Existing 3D datasets are either low resolution, limited to a small amount of scenes, based on images of varying quality because retrieved from the internet, or limited to specific capturing scenarios. Motivated by this lack of suitable 3D datasets, we captured TerraSky3D, a high-resolution large-scale 3D reconstruction dataset comprising 50,000 images divided into 150 ground, aerial, and mixed scenes. The dataset focuses on European landmarks and comes with curated calibration data, camera poses, and depth maps. TerraSky3D tries to answer the need for challenging dataset that can be used to train and evaluate 3D reconstruction-related pipelines.

cs.CV↗

arXiv2026

A Streamlined Attention-Based Network for Descriptor Extraction

We introduce SANDesc, a Streamlined Attention-Based Network for Descriptor extraction that aims to improve on existing architectures for keypoint description. Our descriptor network learns to compute descriptors that improve matching without modifying the underlying keypoint detector. We employ a revised U-Net-like architecture enhanced with Convolutional Block Attention Modules and residual paths, enabling effective local representation while maintaining computational efficiency. We refer to the building blocks of our model as Residual U-Net Blocks with Attention. The model is trained using a modified triplet loss in combination with a curriculum learning-inspired hard negative mining strategy, which improves training stability. Extensive experiments on HPatches, MegaDepth-1500, and the Image Matching Challenge 2021 show that training SANDesc on top of existing keypoint detectors leads to improved results on multiple matching tasks compared to the original keypoint descriptors. At the same time, SANDesc has a model complexity of just 2.4 million parameters. As a further contribution, we introduce a new urban dataset featuring 4K images and pre-calibrated intrinsics, designed to evaluate feature extractors. On this benchmark, SANDesc achieves substantial performance gains over the existing descriptors while operating with limited computational resources.

cs.CV↗

arXiv2025

MultimodalStudio: A Heterogeneous Sensor Dataset and Framework for Neural Rendering across Multiple Imaging Modalities

Neural Radiance Fields (NeRF) have shown impressive performances in the rendering of 3D scenes from arbitrary viewpoints. While RGB images are widely preferred for training volume rendering models, the interest in other radiance modalities is also growing. However, the capability of the underlying implicit neural models to learn and transfer information across heterogeneous imaging modalities has seldom been explored, mostly due to the limited training data availability. For this purpose, we present MultimodalStudio (MMS): it encompasses MMS-DATA and MMS-FW. MMS-DATA is a multimodal multi-view dataset containing 32 scenes acquired with 5 different imaging modalities: RGB, monochrome, near-infrared, polarization and multispectral. MMS-FW is a novel modular multimodal NeRF framework designed to handle multimodal raw data and able to support an arbitrary number of multi-channel devices. Through extensive experiments, we demonstrate that MMS-FW trained on MMS-DATA can transfer information between different imaging modalities and produce higher quality renderings than using single modalities alone. We publicly release the dataset and the framework, to promote the research on multimodal volume rendering and beyond.

cs.GR↗

arXiv2024

Extended Drag-Based Model for better predicting the evolution of Coronal Mass Ejections

The solar wind drag-based model is a widely used framework for predicting the propagation of Coronal Mass Ejections (CMEs) through interplanetary space. This model primarily considers the aerodynamic drag exerted by the solar wind on CMEs. However, factors like magnetic forces, pressure gradients, and the internal dynamics within CMEs justify the need of introducing an additional small-scale acceleration term in the game. Indeed, by accounting for this extra acceleration, the extended drag-based model is shown to offer improved accuracy in describing the evolution of CMEs through the heliosphere and, in turn, in forecasting CME trajectories and arrival times at Earth. This enhancement is crucial for better predicting Space Weather events and mitigating their potential impacts on space-based and terrestrial technologies.

astro-ph.SR↗

arXiv2024

GMM-IKRS: Gaussian Mixture Models for Interpretable Keypoint Refinement and Scoring

The extraction of keypoints in images is at the basis of many computer vision applications, from localization to 3D reconstruction. Keypoints come with a score permitting to rank them according to their quality. While learned keypoints often exhibit better properties than handcrafted ones, their scores are not easily interpretable, making it virtually impossible to compare the quality of individual keypoints across methods. We propose a framework that can refine, and at the same time characterize with an interpretable score, the keypoints extracted by any method. Our approach leverages a modified robust Gaussian Mixture Model fit designed to both reject non-robust keypoints and refine the remaining ones. Our score comprises two components: one relates to the probability of extracting the same keypoint in an image captured from another viewpoint, the other relates to the localization accuracy of the keypoint. These two interpretable components permit a comparison of individual keypoints extracted across different methods. Through extensive experiments we demonstrate that, when applied to popular keypoint detectors, our framework consistently improves the repeatability of keypoints as well as their performance in homography and two/multiple-view pose recovery tasks.

cs.CV↗

arXiv2024

A Kustaanheimo-Stiefel regularization of the elliptic restricted three-body problem and the detection of close encounters with fast Lyapunov indicators

We present the Kustaanheimo-Stiefel (KS) regularization of the elliptic restricted three-body problem (ER3BP) at the secondary body $P_2$, and discuss its use to study a category of transits through its Hill's sphere (fast close encounters). Starting from the Hamiltonian representation of the problem using the synodic rotating-pulsating reference frame and the true anomaly of $P_2$ as independent variable, we perform the regularization at the secondary body analogous to the circular case by applying the classical KS transformation and the iso-energetic reduction in an extended 10-dimensional phase-space. Using such regularized Hamiltonian we recover a definition of fast close encounters in the ER3BP for small values of the mass parameter $μ$ (while we do not require a smallness condition on the eccentricity of the primaries), and we show that for these encounters the solutions of the variational equations are characterized by an exponential growth during the fast transits through the Hill's sphere. Thus, for small $μ$, we justify the effectiveness of the regularized fast Lyapunov indicators (RFLIs) to detect orbits with multiple fast close encounters. Finally, we provide numerical demonstrations and show the benefits of the regularization in terms of the computational cost.

math-ph↗

arXiv2023

Exploiting Multiple Priors for Neural 3D Indoor Reconstruction

Neural implicit modeling permits to achieve impressive 3D reconstruction results on small objects, while it exhibits significant limitations in large indoor scenes. In this work, we propose a novel neural implicit modeling method that leverages multiple regularization strategies to achieve better reconstructions of large indoor environments, while relying only on images. A sparse but accurate depth prior is used to anchor the scene to the initial model. A dense but less accurate depth prior is also introduced, flexible enough to still let the model diverge from it to improve the estimated geometry. Then, a novel self-supervised strategy to regularize the estimated surface normals is presented. Finally, a learnable exposure compensation scheme permits to cope with challenging lighting conditions. Experimental results show that our approach produces state-of-the-art 3D reconstructions in challenging indoor scenarios.

cs.CV↗

arXiv2023

S-TREK: Sequential Translation and Rotation Equivariant Keypoints for local feature extraction

In this work we introduce S-TREK, a novel local feature extractor that combines a deep keypoint detector, which is both translation and rotation equivariant by design, with a lightweight deep descriptor extractor. We train the S-TREK keypoint detector within a framework inspired by reinforcement learning, where we leverage a sequential procedure to maximize a reward directly related to keypoint repeatability. Our descriptor network is trained following a "detect, then describe" approach, where the descriptor loss is evaluated only at those locations where keypoints have been selected by the already trained detector. Extensive experiments on multiple benchmarks confirm the effectiveness of our proposed method, with S-TREK often outperforming other state-of-the-art methods in terms of repeatability and quality of the recovered poses, especially when dealing with in-plane rotations.

cs.CV↗

arXiv2023

Relegation-free closed-form perturbation theory and the domain of secular motions in the Restricted 3-Body Problem

We propose a closed-form (i.e. without expansion in the orbital eccentricities) scheme for computations in perturbation theory in the restricted three-body problem (R3BP) when the massless particle is in an orbit exterior to the one of the primary perturber. Starting with a multipole expansion of the barycentric (Jacobi-reduced) Hamiltonian, we carry out a sequence of normalizations in Delaunay variables by Lie series, leading to a secular Hamiltonian model without use of relegation. To this end, we introduce a book-keeping analogous to the one proposed in Cavallari and Efthymiopoulos (2022) for test particle orbits interior to the one of the primary perturber, but here adapted, instead, to the case of exterior orbits. We give numerical examples of the performance of the method in both the planar circular and the spatial elliptic restricted three-body problem, for parameters pertinent to the Sun-Jupiter system. In particular, we demonstrate the method's accuracy in terms of reproducibility of the orbital elements' variations far from mean-motion resonances. As a basic outcome of the method, we show how, using as criterion the size of the series' remainder, we reach to obtain an accurate semi-analytical estimate of the boundary (in the space of orbital elements) where the secular Hamiltonian model arrived at after eliminating the particle's fast degree of freedom provides a valid approximation of the true dynamics.

math-ph↗

arXiv2022

DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo

We propose a novel approach for deep learning-based Multi-View Stereo (MVS). For each pixel in the reference image, our method leverages a deep architecture to search for the corresponding point in the source image directly along the corresponding epipolar line. We denote our method DELS-MVS: Deep Epipolar Line Search Multi-View Stereo. Previous works in deep MVS select a range of interest within the depth space, discretize it, and sample the epipolar line according to the resulting depth values: this can result in an uneven scanning of the epipolar line, hence of the image space. Instead, our method works directly on the epipolar line: this guarantees an even scanning of the image space and avoids both the need to select a depth range of interest, which is often not known a priori and can vary dramatically from scene to scene, and the need for a suitable discretization of the depth space. In fact, our search is iterative, which avoids the building of a cost volume, costly both to store and to process. Finally, our method performs a robust geometry-aware fusion of the estimated depth maps, leveraging a confidence predicted alongside each depth. We test DELS-MVS on the ETH3D, Tanks and Temples and DTU benchmarks and achieve competitive results with respect to state-of-the-art approaches.

cs.CV↗

arXiv2022

MD-Net: Multi-Detector for Local Feature Extraction

Establishing a sparse set of keypoint correspon dences between images is a fundamental task in many computer vision pipelines. Often, this translates into a computationally expensive nearest neighbor search, where every keypoint descriptor at one image must be compared with all the descriptors at the others. In order to lower the computational cost of the matching phase, we propose a deep feature extraction network capable of detecting a predefined number of complementary sets of keypoints at each image. Since only the descriptors within the same set need to be compared across the different images, the matching phase computational complexity decreases with the number of sets. We train our network to predict the keypoints and compute the corresponding descriptors jointly. In particular, in order to learn complementary sets of keypoints, we introduce a novel unsupervised loss which penalizes intersections among the different sets. Additionally, we propose a novel descriptor-based weighting scheme meant to penalize the detection of keypoints with non-discriminative descriptors. With extensive experiments we show that our feature extraction network, trained only on synthetically warped images and in a fully unsupervised manner, achieves competitive results on 3D reconstruction and re-localization tasks at a reduced matching complexity.

cs.CV↗

arXiv2021

IB-MVS: An Iterative Algorithm for Deep Multi-View Stereo based on Binary Decisions

We present a novel deep-learning-based method for Multi-View Stereo. Our method estimates high resolution and highly precise depth maps iteratively, by traversing the continuous space of feasible depth values at each pixel in a binary decision fashion. The decision process leverages a deep-network architecture: this computes a pixelwise binary mask that establishes whether each pixel actual depth is in front or behind its current iteration individual depth hypothesis. Moreover, in order to handle occluded regions, at each iteration the results from different source images are fused using pixelwise weights estimated by a second network. Thanks to the adopted binary decision strategy, which permits an efficient exploration of the depth space, our method can handle high resolution images without trading resolution and precision. This sets it apart from most alternative learning-based Multi-View Stereo methods, where the explicit discretization of the depth space requires the processing of large cost volumes. We compare our method with state-of-the-art Multi-View Stereo methods on the DTU, Tanks and Temples and the challenging ETH3D benchmarks and show competitive results.

cs.CV↗

arXiv2020

BP-MVSNet: Belief-Propagation-Layers for Multi-View-Stereo

In this work, we propose BP-MVSNet, a convolutional neural network (CNN)-based Multi-View-Stereo (MVS) method that uses a differentiable Conditional Random Field (CRF) layer for regularization. To this end, we propose to extend the BP layer and add what is necessary to successfully use it in the MVS setting. We therefore show how we can calculate a normalization based on the expected 3D error, which we can then use to normalize the label jumps in the CRF. This is required to make the BP layer invariant to different scales in the MVS setting. In order to also enable fractional label jumps, we propose a differentiable interpolation step, which we embed into the computation of the pairwise term. These extensions allow us to integrate the BP layer into a multi-scale MVS network, where we continuously improve a rough initial estimate until we get high quality depth maps as a result. We evaluate the proposed BP-MVSNet in an ablation study and conduct extensive experiments on the DTU, Tanks and Temples and ETH3D data sets. The experiments show that we can significantly outperform the baseline and achieve state-of-the-art results.

cs.CV↗

arXiv2020

Joint Graph-based Depth Refinement and Normal Estimation

Depth estimation is an essential component in understanding the 3D geometry of a scene, with numerous applications in urban and indoor settings. These scenes are characterized by a prevalence of human made structures, which in most of the cases, are either inherently piece-wise planar, or can be approximated as such. In these settings, we devise a novel depth refinement framework that aims at recovering the underlying piece-wise planarity of the inverse depth map. We formulate this task as an optimization problem involving a data fidelity term that minimizes the distance to the input inverse depth map, as well as a regularization that enforces a piece-wise planar solution. As for the regularization term, we model the inverse depth map as a weighted graph between pixels. The proposed regularization is designed to estimate a plane automatically at each pixel, without any need for an a priori estimation of the scene planes, and at the same time it encourages similar pixels to be assigned to the same plane. The resulting optimization problem is efficiently solved with ADAM algorithm. Experiments show that our method leads to a significant improvement in depth refinement, both visually and numerically, with respect to state-of-the-art algorithms on Middlebury, KITTI and ETH3D multi-view stereo datasets.

cs.CV↗

arXiv2020

DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction

Deep Neural Networks (DNNs) have the potential to improve the quality of image-based 3D reconstructions. However, the use of DNNs in the context of 3D reconstruction from large and high-resolution image datasets is still an open challenge, due to memory and computational constraints. We propose a pipeline which takes advantage of DNNs to improve the quality of 3D reconstructions while being able to handle large and high-resolution datasets. In particular, we propose a confidence prediction network explicitly tailored for Multi-View Stereo (MVS) and we use it for both depth map outlier filtering and depth map refinement within our pipeline, in order to improve the quality of the final 3D reconstructions. We train our confidence prediction network on (semi-)dense ground truth depth maps from publicly available real world MVS datasets. With extensive experiments on popular benchmarks, we show that our overall pipeline can produce state-of-the-art 3D reconstructions, both qualitatively and quantitatively.

cs.CV↗