Isaac Corley

November 2025

Country-Scale Agricultural Field Boundary Predictions

Through my collaboration with Wherobots & Taylor Geospatial Engine, we have open-sourced planting/harvest season mosaics and field boundary predictions for 5 countries for 2023 and 2024.

October 2025

Spatial Stack Podcast: Beyond the Hype: Embeddings, Foundation Models, and the Future of Earth Observation

I joined Matt Forrest's Spatial Stack podcast with Chris Ren to discuss the current state of Geospatial Foundation Models and Embeddings.

August 2025

Satellite-Image-Deep-Learning Podcast: Chained Models for High-Res Aerial Solar Fault Detection

I joined Robin Cole's Satellite-Image-Deep-Learning podcast to discuss our CPVR PBVS paper: Aerial Infrared Health Monitoring of Solar Photovoltaic Farms at Scale.

HydroChronos: Forecasting Decades of Surface Water Change

Daniele Rege Cambrin, Eleonora Poeta, Eliana Pastor, Isaac Corley, Tania Cerquitelli, Elena Baralis, Paolo Garza

We introduce HydroChronos, a large-scale multi-modal dataset of over three decades of satellite imagery & climate data for forecasting long-term surface water dynamics.

InspectVLM: Unified in Theory, Unreliable in Practice

Conor Wallace, Isaac Corley, Jonathan Lwowski

We present the InspectMM benchmark, a suite of aerial inspection vision tasks for evaluating VLM multi-tasking capabilities.

Landsat-Bench: Datasets and Benchmarks for Landsat Foundation Models

Isaac Corley, Lakshay Sharma, Ruth Crasto

This paper introduces Landsat-Bench, a set of benchmarks to evaluate Landsat-based foundation models, showing that pretraining on SSL4EO-L outperforms ImageNet for remote sensing tasks.

Aerial Infrared Health Monitoring of Solar Photovoltaic Farms at Scale

Isaac Corley, Conor Wallace, Sourav Agrawal, Burton Putrah, Jonathan Lwowski

We collect a large-scale aerial infrared dataset of 6k+ solar farms across the continental U.S. and detail our end-to-end pipeline for detecting anomalies and estimating plant-level power loss due to defective panels.

FLAVARS: A Multimodal Foundational Language and Vision Alignment Model for Remote Sensing

Isaac Corley, Simone Fobi Nsutezo, Anthony Ortiz, Caleb Robinson, Rahul Dodhia, Juan M. Lavista Ferres, Peyman Najafirad

We find that pretraining using CLIP+MAE+MLM+SatCLIP objectives provides better balance for dense vision tasks over pure CLIP and FLAVA pretraining.

A Change Detection Reality Check

Isaac Corley, Caleb Robinson, Anthony Ortiz

We find a U-Net baseline (2015) is still a top performer on change detection benchmarks.

Barely-Visible Surface Crack Detection for Wind Turbine Sustainability

Sourav Agrawal, Isaac Corley, Conor Wallace, Clovis Vaughn, Jonathan Lwowski

We present a novel dataset and pipeline for detecting barely-visible surface cracks on wind turbine blades.

Depth Any Canopy: Leveraging Depth Foundation Models for Canopy Height Estimation

Daniele Rege Cambrin, Isaac Corley, Paolo Garza

We efficiently adapt sota monocular depth estimation models for tree canopy height estimation in aerial & satellite imagery.

ZRG: A Dataset for Multimodal 3D Residential Rooftop Understanding

Isaac Corley, Jonathan Lwowski, Peyman Najafirad

We present a novel large-scale dataset for 3D understanding of residential roofs using orthomosaics, DSMs, and 3D roof wireframes.

Seeing the roads through the trees: A benchmark for modeling spatial dependencies with aerial imagery

Caleb Robinson, Isaac Corley, Anthony Ortiz, Rahul Dodhia, Juan M. Lavista Ferres, Peyman Najafirad

We introduce a novel dataset for evaluating a model's ability to use long-range spatial context by performing road extraction in aerial imagery with high amounts of occlusion by tree canopy.

Revisiting Pre-trained Remote Sensing Model Benchmarks: Resizing and Normalization Matters

Isaac Corley, Caleb Robinson, Rahul Dodhia, Juan M. Lavista Ferres, Peyman Najafirad

We perform a comprehensive benchmark of geospatial foundation models and find that they are highly sensitive to pretrained image size and normalization.

SSL4EO-L: Datasets and Foundation Models for Landsat Imagery

Adam J. Stewart, Nils Lehmann, Isaac A. Corley, Yi Wang, Yi-Chia Chang, Nassim Ait Ali Braham, Shradha Sehgal, Caleb Robinson, Arindam Banerjee

We introduce SSL4EO-L, the first ever dataset designed for self-supervised learning for Earth Observation for the Landsat family of satellites.

TorchGeo: Deep Learning with Geospatial Data

Adam J. Stewart, Caleb Robinson, Isaac A. Corley, Anthony Ortiz, Juan M. Lavista Ferres, Arindam Banerjee

We introduce TorchGeo, a Python library for integrating geospatial data into the PyTorch deep learning ecosystem.

Supervising Remote Sensing Change Detection Models with 3D Surface Semantics

Isaac Corley, Peyman Najafirad

We propose Contrastive Surface-Image Pretraining (CSIP) for joint learning a latent space which extracts surface level features from optical RGB imagery.