Enrico Magli
YOU?
Author Swipe
View article: Deep Lidar-Guided Image Deblurring
Deep Lidar-Guided Image Deblurring Open
The rise in portable Lidar instruments enables new opportunities for depth-assisted image processing. In this paper, we study whether the depth information provided by mobile Lidar sensors present in recent smartphones is useful for the ta…
View article: Onboard Hyperspectral Super-Resolution with Deep Pushbroom Neural Network
Onboard Hyperspectral Super-Resolution with Deep Pushbroom Neural Network Open
Hyperspectral imagers on satellites obtain the fine spectral signatures that are essential in distinguishing one material from another but at the expense of a limited spatial resolution. Enhancing the latter is thus a desirable preprocessi…
View article: First Laboratory Measurements of a Super-Resolved Compressive Instrument in the Medium Infrared
First Laboratory Measurements of a Super-Resolved Compressive Instrument in the Medium Infrared Open
View article: In-flight radiometric calibration of the Metis UV H I Ly-<i>α</i> channel and comparison with UVCS data
In-flight radiometric calibration of the Metis UV H I Ly-<i>α</i> channel and comparison with UVCS data Open
Context. We present the results of the in-flight radiometric calibration performed for the ultraviolet (UV) H I Ly- α channel of the Metis coronagraph on board Solar Orbiter. Aims. The radiometric calibration is a fundamental procedure req…
View article: Multitask AI for onboard intelligence in Earth observation missions
Multitask AI for onboard intelligence in Earth observation missions Open
View article: Deep Lidar-guided Image Deblurring
Deep Lidar-guided Image Deblurring Open
The rise of portable Lidar instruments, including their adoption in smartphones, opens the door to novel computational imaging techniques. Being an active sensing instrument, Lidar can provide complementary data to passive optical sensors,…
View article: DreamCache: Finetuning-Free Lightweight Personalized Image Generation via Feature Caching
DreamCache: Finetuning-Free Lightweight Personalized Image Generation via Feature Caching Open
Personalized image generation requires text-to-image generative models that capture the core features of a reference subject to allow for controlled generation across different contexts. Existing methods face challenges due to complex trai…
View article: Efficient Onboard Multitask AI Architecture Based on Self-Supervised Learning
Efficient Onboard Multitask AI Architecture Based on Self-Supervised Learning Open
There is growing interest toward the use of artificial intelligence (AI) directly onboard satellites for quick analysis and rapid response to critical events such as natural disasters. This article presents a blueprint to the mission desig…
View article: Metis Observation of the Onset of Fully Developed Turbulence in the Solar Corona
Metis Observation of the Onset of Fully Developed Turbulence in the Solar Corona Open
This Letter reports the first observation of the onset of fully developed turbulence in the solar corona. Long time series of white-light coronal images, acquired by Metis aboard Solar Orbiter at 2 minutes cadence and spanning about 10 hr,…
View article: Efficient onboard multi-task AI architecture based on self-supervised learning
Efficient onboard multi-task AI architecture based on self-supervised learning Open
There is growing interest towards the use of AI directly onboard satellites for quick analysis and rapid response to critical events such as natural disasters. This paper presents a blueprint to the mission designer for the development of …
View article: MotionCraft: Physics-based Zero-Shot Video Generation
MotionCraft: Physics-based Zero-Shot Video Generation Open
Generating videos with realistic and physically plausible motion is one of the main recent challenges in computer vision. While diffusion models are achieving compelling results in image generation, video diffusion models are limited by he…
View article: Modeling uncertainty for Gaussian Splatting
Modeling uncertainty for Gaussian Splatting Open
We present Stochastic Gaussian Splatting (SGS): the first framework for uncertainty estimation using Gaussian Splatting (GS). GS recently advanced the novel-view synthesis field by achieving impressive reconstruction quality at a fraction …
View article: Onboard deep lossless and near-lossless predictive coding of hyperspectral images with line-based attention
Onboard deep lossless and near-lossless predictive coding of hyperspectral images with line-based attention Open
Deep learning methods have traditionally been difficult to apply to compression of hyperspectral images onboard of spacecrafts, due to the large computational complexity needed to achieve adequate representational power, as well as the lac…
View article: Cancelable templates for secure face verification based on deep learning and random projections
Cancelable templates for secure face verification based on deep learning and random projections Open
Recently, biometric recognition has become a significant field of research. The concept of cancelable biometrics (CB) has been introduced to address security concerns related to the handling of sensitive data. In this paper, we address unc…
View article: Deep 3D World Models for Multi-Image Super-Resolution Beyond Optical Flow
Deep 3D World Models for Multi-Image Super-Resolution Beyond Optical Flow Open
Multi-image super-resolution (MISR) allows to increase the spatial resolution of a low-resolution (LR) acquisition by combining multiple images carrying complementary information in the form of sub-pixel offsets in the scene sampling, and …
View article: Fast Inference in Denoising Diffusion Models via MMD Finetuning
Fast Inference in Denoising Diffusion Models via MMD Finetuning Open
Denoising Diffusion Models (DDMs) have become a popular tool for generating high-quality samples from complex data distributions. These models are able to capture sophisticated patterns and structures in the data, and can generate samples …
View article: Deep 3D World Models for Multi-Image Super-Resolution Beyond Optical Flow
Deep 3D World Models for Multi-Image Super-Resolution Beyond Optical Flow Open
Multi-image super-resolution (MISR) allows to increase the spatial resolution of a low-resolution (LR) acquisition by combining multiple images carrying complementary information in the form of sub-pixel offsets in the scene sampling, and …
View article: Synthetic Data Pretraining for Hyperspectral Image Super-Resolution
Synthetic Data Pretraining for Hyperspectral Image Super-Resolution Open
Large-scale self-supervised pretraining of deep learning models is known to be critical in several fields, such as language processing, where its has led to significant breakthroughs. Indeed, it is often more impactful than architectural d…
View article: Data Simulations of a Compressive Sensing Multispectral Imager in the Mid-Infrared Region and Its Performances for the Monitoring of High-Temperature Events
Data Simulations of a Compressive Sensing Multispectral Imager in the Mid-Infrared Region and Its Performances for the Monitoring of High-Temperature Events Open
The mid-infrared spectral (MIR) region is poorly exploited in Earth Observation (EO) applications despite its potential for impacting several application fields, from climatological studies to land management. Among these, high-temperature…
View article: Exploring the Potential of Compressive Sensing Payloads for Earth Observation from Geostationary Platforms: An Instrumental Concept for Fire Monitoring
Exploring the Potential of Compressive Sensing Payloads for Earth Observation from Geostationary Platforms: An Instrumental Concept for Fire Monitoring Open
Earth observation (EO) payload performances in the infrared spectral region from geostationary platforms are often limited by spatial resolution. In this paper, we investigate an instrumental concept leveraging a compressive sensing paradi…
View article: Coronal Heating Rate in the Slow Solar Wind
Coronal Heating Rate in the Slow Solar Wind Open
This Letter reports the first observational estimate of the heating rate in the slowly expanding solar corona. The analysis exploits the simultaneous remote and local observations of the same coronal plasma volume, with the Solar Orbiter/M…
View article: Prediction Capability of Geomagnetic Events from Solar Wind Data Using Neural Networks
Prediction Capability of Geomagnetic Events from Solar Wind Data Using Neural Networks Open
Multiple neural network architectures, with different structural composition and complexity, are implemented in this study with the aim of providing multi-hour-ahead warnings of severe geomagnetic disturbances, based on in situ measurement…
View article: Super-resolved compressive demonstrator for Earth Observation applications in the Medium Infrared: instrumental concept, optical design and expected performances
Super-resolved compressive demonstrator for Earth Observation applications in the Medium Infrared: instrumental concept, optical design and expected performances Open
Earth Observation (EO) systems are generating an ever-increasing amount of data to be handled on board yet with limited resources, which sometimes hinders a full exploitation of the information content. In this paper, we present a demonstr…
View article: Coronal Heating Rate in the Slow Solar Wind
Coronal Heating Rate in the Slow Solar Wind Open
This Letter reports the first observational estimate of the heating rate in the slowly expanding solar corona. The analysis exploits the simultaneous remote and local observations of the same coronal plasma volume with the Solar Orbiter/Me…
View article: In-flight radiometric calibration of the Metis Visible Light channel using stars and comparison with STEREO-A/COR2 data
In-flight radiometric calibration of the Metis Visible Light channel using stars and comparison with STEREO-A/COR2 data Open
Context. We present the results for the in-flight radiometric calibration performed for the Visible Light (VL) channel of the Metis coronagraph on board Solar Orbiter. Aims. The radiometric calibration is a fundamental step in building the…
View article: Multimedia Signal Processing: A history of the Multimedia Signal Processing Technical Committee
Multimedia Signal Processing: A history of the Multimedia Signal Processing Technical Committee Open
International audience
View article: Gaussian class-conditional simplex loss for accurate, adversarially robust deep classifier training
Gaussian class-conditional simplex loss for accurate, adversarially robust deep classifier training Open
In this work, we present the Gaussian Class-Conditional Simplex (GCCS) loss: a novel approach for training deep robust multiclass classifiers that improves over the state-of-the-art in terms of classification accuracy and adversarial robus…
View article: Compressive Sensing Imaging Spectrometer for UV-Vis Stellar Spectroscopy: Instrumental Concept and Performance Analysis
Compressive Sensing Imaging Spectrometer for UV-Vis Stellar Spectroscopy: Instrumental Concept and Performance Analysis Open
Compressive sensing (CS) has been proposed as a disruptive approach to developing a novel class of optical instrumentation used in diverse application domains. Thanks to sparsity as an inherent feature of many natural signals, CS allows fo…
View article: Fast Inference in Denoising Diffusion Models via MMD Finetuning
Fast Inference in Denoising Diffusion Models via MMD Finetuning Open
Denoising Diffusion Models (DDMs) have become a popular tool for generating high-quality samples from complex data distributions. These models are able to capture sophisticated patterns and structures in the data, and can generate samples …
View article: Deep Learning Methods for Satellite Compressive Imaging
Deep Learning Methods for Satellite Compressive Imaging Open
Session 5B-Prette-Deep Learning Methods for Satellite Compressive Imaging