Image and Signal Processing • ISP
@isp-uv-es.bsky.social
#Artificial #Intelligence for #Earth, #Brain, #Social and #Climate sciences • Universitat de València @uv.es
More papers from our team ➡️ http://linktr.ee/ISPUV
More papers from our team ➡️ http://linktr.ee/ISPUV
🥳🎓 Many congrats to Pablo Hernández Cámara, one of our PhD students, who successfully defended his thesis today: "Perceptual Alignment in Artificial Vision: Bio-Inspired Design and Psychophysical Evaluation"!
🧵
🧵
November 14, 2025 at 12:08 PM
🥳🎓 Many congrats to Pablo Hernández Cámara, one of our PhD students, who successfully defended his thesis today: "Perceptual Alignment in Artificial Vision: Bio-Inspired Design and Psychophysical Evaluation"!
🧵
🧵
✍️ Register for the Joint Winter School “AI for Earth System, Hazards & Climate Extremes” here tr.ee/RegistrationWinterSchool2026
📅 March 16 – 20 2026
📍 Venue, NTUA, Athens, Greece
ℹ️ More info at tr.ee/JointWinterSchool2026
@ai4pex.bsky.social @bgc-jena.mpg.de @ellisunitjena.bsky.social @ecmwf.int
📅 March 16 – 20 2026
📍 Venue, NTUA, Athens, Greece
ℹ️ More info at tr.ee/JointWinterSchool2026
@ai4pex.bsky.social @bgc-jena.mpg.de @ellisunitjena.bsky.social @ecmwf.int
November 13, 2025 at 9:15 AM
✍️ Register for the Joint Winter School “AI for Earth System, Hazards & Climate Extremes” here tr.ee/RegistrationWinterSchool2026
📅 March 16 – 20 2026
📍 Venue, NTUA, Athens, Greece
ℹ️ More info at tr.ee/JointWinterSchool2026
@ai4pex.bsky.social @bgc-jena.mpg.de @ellisunitjena.bsky.social @ecmwf.int
📅 March 16 – 20 2026
📍 Venue, NTUA, Athens, Greece
ℹ️ More info at tr.ee/JointWinterSchool2026
@ai4pex.bsky.social @bgc-jena.mpg.de @ellisunitjena.bsky.social @ecmwf.int
🔎🌏 Feini Huang is a postdoc researcher specialising in applying and evaluating Explainable AI (XAI), will use causal inference and spatio-temporal deep learning, to deepen our understanding of Earth systems.
November 4, 2025 at 10:03 AM
🔎🌏 Feini Huang is a postdoc researcher specialising in applying and evaluating Explainable AI (XAI), will use causal inference and spatio-temporal deep learning, to deepen our understanding of Earth systems.
🤖🧠 Kexuan Zhang will continue her research on machine learning, causal inference, and applications in climate analysis.
💻📊 Paolo Frazzetto will focus on developing Graph Neural Networks to address Earth System science problems, blending principles of physics-aware ML, causality, and trustworthy AI.
💻📊 Paolo Frazzetto will focus on developing Graph Neural Networks to address Earth System science problems, blending principles of physics-aware ML, causality, and trustworthy AI.
November 4, 2025 at 10:03 AM
🤖🧠 Kexuan Zhang will continue her research on machine learning, causal inference, and applications in climate analysis.
💻📊 Paolo Frazzetto will focus on developing Graph Neural Networks to address Earth System science problems, blending principles of physics-aware ML, causality, and trustworthy AI.
💻📊 Paolo Frazzetto will focus on developing Graph Neural Networks to address Earth System science problems, blending principles of physics-aware ML, causality, and trustworthy AI.
🌏➗ Carmen Oliver will focus her research on developing multimodal foundation models for Earth modelling.
🗺️🌧️ Paula Costa will be focused on developing risk assessment maps to better understand the impacts of climate extremes and support evidence-based decision-making.
🗺️🌧️ Paula Costa will be focused on developing risk assessment maps to better understand the impacts of climate extremes and support evidence-based decision-making.
November 4, 2025 at 10:03 AM
🌏➗ Carmen Oliver will focus her research on developing multimodal foundation models for Earth modelling.
🗺️🌧️ Paula Costa will be focused on developing risk assessment maps to better understand the impacts of climate extremes and support evidence-based decision-making.
🗺️🌧️ Paula Costa will be focused on developing risk assessment maps to better understand the impacts of climate extremes and support evidence-based decision-making.
👋 Pleased to welcome our new PhD students and postdoc researchers: Carmen Oliver (@elliot-eu.bsky.social), Paula Costa (MeDiTwin Project), Kexuan Zhang, Paolo Frazzetto (Thinking Earth project), and Feini Huang.
👏 We're happy you are joining the ISP team!
👏 We're happy you are joining the ISP team!
November 4, 2025 at 10:03 AM
👋 Pleased to welcome our new PhD students and postdoc researchers: Carmen Oliver (@elliot-eu.bsky.social), Paula Costa (MeDiTwin Project), Kexuan Zhang, Paolo Frazzetto (Thinking Earth project), and Feini Huang.
👏 We're happy you are joining the ISP team!
👏 We're happy you are joining the ISP team!
➡️ Last week, part of our team delivered a series of courses in AI, explainability and modelling for environmental research as part of the MeDiTwin Project.
👥👏 Thanks to the Remote Sensing Lab of the National Technical University of Athens for organising and hosting these courses!
👥👏 Thanks to the Remote Sensing Lab of the National Technical University of Athens for organising and hosting these courses!
October 27, 2025 at 10:45 AM
➡️ Last week, part of our team delivered a series of courses in AI, explainability and modelling for environmental research as part of the MeDiTwin Project.
👥👏 Thanks to the Remote Sensing Lab of the National Technical University of Athens for organising and hosting these courses!
👥👏 Thanks to the Remote Sensing Lab of the National Technical University of Athens for organising and hosting these courses!
🎓 We’re proud to announce that Jordi Cortés Andrés, one of our PhD students, just defended his thesis: "Regularization in Machine Learning: Advances for Earth Sciences".
🥳 Congratulations on reaching this important milestone in your research journey!
🥳 Congratulations on reaching this important milestone in your research journey!
October 16, 2025 at 3:59 PM
🎓 We’re proud to announce that Jordi Cortés Andrés, one of our PhD students, just defended his thesis: "Regularization in Machine Learning: Advances for Earth Sciences".
🥳 Congratulations on reaching this important milestone in your research journey!
🥳 Congratulations on reaching this important milestone in your research journey!
👩🏫 This past week, our colleague @mariapiles.bsky.social was involved in the ceiA3 International Agri-Food Campus of Excellence, where she offered a course on passive microwave remote sensing of soil moisture.
October 15, 2025 at 9:10 AM
👩🏫 This past week, our colleague @mariapiles.bsky.social was involved in the ceiA3 International Agri-Food Campus of Excellence, where she offered a course on passive microwave remote sensing of soil moisture.
⏮️ Last week some of our colleagues attended the CIMR L2PAD workshop organized by the @icmcsic.bsky.social - CSIC.
👥 The activity brought together the community around open Level-2 algorithms and the Ground Processing Prototype for @esa.int's Copernicus CIMR mission.
👥 The activity brought together the community around open Level-2 algorithms and the Ground Processing Prototype for @esa.int's Copernicus CIMR mission.
October 8, 2025 at 10:17 AM
⏮️ Last week some of our colleagues attended the CIMR L2PAD workshop organized by the @icmcsic.bsky.social - CSIC.
👥 The activity brought together the community around open Level-2 algorithms and the Ground Processing Prototype for @esa.int's Copernicus CIMR mission.
👥 The activity brought together the community around open Level-2 algorithms and the Ground Processing Prototype for @esa.int's Copernicus CIMR mission.
🛰️🧑🏫 Last week, our colleagues Luis Gómez-Chova and Cesar Aybar attended the 4th IADF School and offered a workshop diving into trustworthy super-resolution for multispectral remote sensing images.
👐 Thanks to the IEEE Geoscience and Remote Sensing Society for the opportunity!
#EarthObservation
👐 Thanks to the IEEE Geoscience and Remote Sensing Society for the opportunity!
#EarthObservation
September 24, 2025 at 3:04 PM
🛰️🧑🏫 Last week, our colleagues Luis Gómez-Chova and Cesar Aybar attended the 4th IADF School and offered a workshop diving into trustworthy super-resolution for multispectral remote sensing images.
👐 Thanks to the IEEE Geoscience and Remote Sensing Society for the opportunity!
#EarthObservation
👐 Thanks to the IEEE Geoscience and Remote Sensing Society for the opportunity!
#EarthObservation
👥 This week, we welcomed Tobias Braun, from the Institute for Earth System Science & Remote Sensing (Leipzig University).
🌏 During his talk today, he discussed the potential of complexity science to bridge the gap between synoptic processes and planetary-scale dynamics.
🙌 Thanks for visiting us!
🌏 During his talk today, he discussed the potential of complexity science to bridge the gap between synoptic processes and planetary-scale dynamics.
🙌 Thanks for visiting us!
September 19, 2025 at 3:03 PM
👥 This week, we welcomed Tobias Braun, from the Institute for Earth System Science & Remote Sensing (Leipzig University).
🌏 During his talk today, he discussed the potential of complexity science to bridge the gap between synoptic processes and planetary-scale dynamics.
🙌 Thanks for visiting us!
🌏 During his talk today, he discussed the potential of complexity science to bridge the gap between synoptic processes and planetary-scale dynamics.
🙌 Thanks for visiting us!
🆕 New paper out dissecting evapotranspiration (ET) dynamics across Central Europe under varying moisture conditions to better understand how drought impacts plant–atmosphere water fluxes.
🔗 Read it here: https://tr.ee/AssessingEvapotranspirationDynamics
#ClimateScience #Drought
🔗 Read it here: https://tr.ee/AssessingEvapotranspirationDynamics
#ClimateScience #Drought
September 15, 2025 at 1:03 PM
🆕 New paper out dissecting evapotranspiration (ET) dynamics across Central Europe under varying moisture conditions to better understand how drought impacts plant–atmosphere water fluxes.
🔗 Read it here: https://tr.ee/AssessingEvapotranspirationDynamics
#ClimateScience #Drought
🔗 Read it here: https://tr.ee/AssessingEvapotranspirationDynamics
#ClimateScience #Drought
🌱 This calibration improves the accuracy of soil moisture estimates when evaluated against ground networks and model simulations (ECMWF), while also enhancing the spatial and temporal consistency of vegetation optical depth (L-VOD) with reference biomass and vegetation indices.
September 10, 2025 at 3:39 PM
🌱 This calibration improves the accuracy of soil moisture estimates when evaluated against ground networks and model simulations (ECMWF), while also enhancing the spatial and temporal consistency of vegetation optical depth (L-VOD) with reference biomass and vegetation indices.
📊 The study replaces empirical approximations with a machine learning model trained on soil and terrain properties, producing a physically based global map of roughness that better captures surface heterogeneity.
September 10, 2025 at 3:39 PM
📊 The study replaces empirical approximations with a machine learning model trained on soil and terrain properties, producing a physically based global map of roughness that better captures surface heterogeneity.
🆕 🗞️ New paper presenting a new global calibration of soil surface roughness for the SMOS-IC algorithm, which retrieves soil moisture and vegetation optical depth from passive L-band microwave satellite observations.
🔗 Read it here: tr.ee/NovelCalibrationoGlobalSoilRoughness
#RemoteSensing
🔗 Read it here: tr.ee/NovelCalibrationoGlobalSoilRoughness
#RemoteSensing
September 10, 2025 at 3:39 PM
🆕 🗞️ New paper presenting a new global calibration of soil surface roughness for the SMOS-IC algorithm, which retrieves soil moisture and vegetation optical depth from passive L-band microwave satellite observations.
🔗 Read it here: tr.ee/NovelCalibrationoGlobalSoilRoughness
#RemoteSensing
🔗 Read it here: tr.ee/NovelCalibrationoGlobalSoilRoughness
#RemoteSensing
🔹Nuria Alabau-Bosque sharing her work on "Hues and Cues: Human vs. CLIP"
🔗 tr.ee/HuesandCuesHumanvsCLIP
#CCN2025
🔗 tr.ee/HuesandCuesHumanvsCLIP
#CCN2025
August 28, 2025 at 3:19 PM
🔹Nuria Alabau-Bosque sharing her work on "Hues and Cues: Human vs. CLIP"
🔗 tr.ee/HuesandCuesHumanvsCLIP
#CCN2025
🔗 tr.ee/HuesandCuesHumanvsCLIP
#CCN2025
🔹Paula Daudén-Oliver with her poster "Response to Affine Transforms of Image Distance Metrics and Humans"
🔗 tr.ee/ResponsetoAffineTransforms
#CCN2025
🔗 tr.ee/ResponsetoAffineTransforms
#CCN2025
August 28, 2025 at 3:19 PM
🔹Paula Daudén-Oliver with her poster "Response to Affine Transforms of Image Distance Metrics and Humans"
🔗 tr.ee/ResponsetoAffineTransforms
#CCN2025
🔗 tr.ee/ResponsetoAffineTransforms
#CCN2025
🔹 "Contrast Sensitivity Function of Multimodal Vision-Language Models" (Alexandra Gomez-Villa)
🔗 tr.ee/ContrastSensitivityFunction
#CCN2025
🔗 tr.ee/ContrastSensitivityFunction
#CCN2025
August 28, 2025 at 3:19 PM
🔹 "Contrast Sensitivity Function of Multimodal Vision-Language Models" (Alexandra Gomez-Villa)
🔗 tr.ee/ContrastSensitivityFunction
#CCN2025
🔗 tr.ee/ContrastSensitivityFunction
#CCN2025
🔹 Pablo Hernández-Cámara sharing his poster on "Evolution of Low-Level and Texture Human-CLIP Alignment" and "From Images to Perception: Emergence of Perceptual Properties by Reconstructing Images"
🔗 tr.ee/EvolutionofLow-LevelandTextureHuman
🔗 tr.ee/FromImagestoPerception
#CCN2025
🔗 tr.ee/EvolutionofLow-LevelandTextureHuman
🔗 tr.ee/FromImagestoPerception
#CCN2025
August 28, 2025 at 3:19 PM
🔹 Pablo Hernández-Cámara sharing his poster on "Evolution of Low-Level and Texture Human-CLIP Alignment" and "From Images to Perception: Emergence of Perceptual Properties by Reconstructing Images"
🔗 tr.ee/EvolutionofLow-LevelandTextureHuman
🔗 tr.ee/FromImagestoPerception
#CCN2025
🔗 tr.ee/EvolutionofLow-LevelandTextureHuman
🔗 tr.ee/FromImagestoPerception
#CCN2025
🔹"Do Vision Transformers See Like Humans? Evaluating their Perceptual Alignment" (Jose Manuel Jaén-Lorites)
🔗 tr.ee/DoVisionTransformersSeeLikeHumans
#CCN2025
🔗 tr.ee/DoVisionTransformersSeeLikeHumans
#CCN2025
August 28, 2025 at 3:19 PM
🔹"Do Vision Transformers See Like Humans? Evaluating their Perceptual Alignment" (Jose Manuel Jaén-Lorites)
🔗 tr.ee/DoVisionTransformersSeeLikeHumans
#CCN2025
🔗 tr.ee/DoVisionTransformersSeeLikeHumans
#CCN2025
🧠 Some of our colleagues presented their work at #CCN2025 this month!
👥 This event brings together researchers in cognitive science, neuroscience, and AI to explore the computations behind complex behaviour.
🥳 Proud to see our team contributing to these conversations!
🧵
👥 This event brings together researchers in cognitive science, neuroscience, and AI to explore the computations behind complex behaviour.
🥳 Proud to see our team contributing to these conversations!
🧵
August 28, 2025 at 3:19 PM
🧠 Some of our colleagues presented their work at #CCN2025 this month!
👥 This event brings together researchers in cognitive science, neuroscience, and AI to explore the computations behind complex behaviour.
🥳 Proud to see our team contributing to these conversations!
🧵
👥 This event brings together researchers in cognitive science, neuroscience, and AI to explore the computations behind complex behaviour.
🥳 Proud to see our team contributing to these conversations!
🧵
📊 Our findings are particularly valuable for operational flood monitoring in data-scarce regions or areas with high environmental variability, where developing robust methods is most challenging.
August 7, 2025 at 8:32 AM
📊 Our findings are particularly valuable for operational flood monitoring in data-scarce regions or areas with high environmental variability, where developing robust methods is most challenging.
🛰️ We developed a dual-stream multimodal architecture that seamlessly integrates both SAR (Sentinel-1) and optical (Sentinel-2) data, enabling comprehensive evaluation across single and multi-modal scenarios.
August 7, 2025 at 8:32 AM
🛰️ We developed a dual-stream multimodal architecture that seamlessly integrates both SAR (Sentinel-1) and optical (Sentinel-2) data, enabling comprehensive evaluation across single and multi-modal scenarios.
🔍 Even high-performing flood detection models trained on single datasets often fail when applied to new regions or different sensor conditions. However, we demonstrate that multi-dataset training strategies significantly enhance model robustness and generalization capabilities.
August 7, 2025 at 8:32 AM
🔍 Even high-performing flood detection models trained on single datasets often fail when applied to new regions or different sensor conditions. However, we demonstrate that multi-dataset training strategies significantly enhance model robustness and generalization capabilities.