University of Texas at San Antonio
Ph. D. in Electrical Engineering, advised by Paul Rad (Peyman Najafirad).
Dissertation: Multimodal Vision for Mapping in Remote Sensing
Fall 2024
University of Texas at San Antonio
M.S. in Electrical Engineering, advised by Yufei Huang.
Thesis: Deep Learning Super-Resolution for EEG Spatial Interpolation
Texas A&M University - Kingsville
B.S. in Electrical Engineering, advised by Lifford McLauchlan.
Minors in Mathematic and Security Engineering.


Zeitview (formerly DroneBase)
Senior Machine Learning Engineer. I develop and deploy computer vision and deep learning models for enhancing renewable energy inspections and analytics, including solar farms, wind turbines, and commercial and residential rooftops.
2021 - 2024
San Antonio, TX
SLB (formerly Schlumberger)
Machine Learning Researcher on the Artificial Intelligence Hub team. I researched using deep learning to estimate global xCO2 using in situ ODIAC Fossil fuel emissions and OCO-2 and GOSAT-2 datasets.
2022 - 2023
San Antonio, TX
Senior Machine Learning Engineer. I applied state-of-the-art Optical Character Recognition (OCR) and Text Summarization methods to parse real estate and financial documents.
2021 - 2022
San Antonio, TX
Senior Computer Vision Engineer. Developed and deployed models to drive the Spectra AI platform's satellite image analytics as well as served as the PI and lead engineer on the IARPA SMART program
2021 - 2022
San Antonio, TX
Senior Data Scientist. Developed and deployed computer vision models for extracting insights and features from real estate property images for improving HouseCanary's Automated Valuation Model (AVM) and property recommender system utilized by real estate investors.
2019 - 2020
San Antonio, TX
Booz Allen Hamilton
Senior Data Scientist on the AI Cybersecurity team. I researched and developed prototypes for deep learning based image steganography detection and removal as well as adversarial domain generation detection.
2018 - 2019
San Antonio, TX
Southwest Research Institute (SwRI)
Research Engineer in the Defense & Intelligence Solutions Division. I developed and deployed software updates to the A-10 Warthog aircraft as well as researched machine learning methods for detecting engine stalls and exploiting the MIL-STD-1553 communications bus while working under Kenneth Holladay.
2016 - 2018
San Antonio, TX
Oak Ridge National Laboratory (ORNL)
Research Intern in the Intelligent Systems Group. I recorded and annotated a dataset of seismic signals of human and vehicle activity and trained machine learning methods to detect this activity.
Oak Ridge, TN


A Change Detection Reality Check
I. Corley, C. Robinson, A. Ortiz
We present an analysis of the current state-of-the-art in change detection literature. We find that a simple baseline of U-Net, an architecture from 2015, is still a top performer on several benchmarks and consistently outperforms many recently proposed methods.
ICLR 2024 Machine Learning for Remote Sensing (ML4RS) Workshop
Seeing the roads through the trees: A benchmark for modeling spatial dependencies with aerial imagery We introduce a novel remote sensing dataset for evaluating a model's ability to learn long-range spatial dependencies in aerial imagery by performing road extraction while containing large gaps occluded by tree canopy.
ArXiv 2024
ZRG: A Dataset for Multimodal 3D Residential Rooftop Understanding
I. Corley, J. Lwowski, P. Najafirad
We present a novel dataset for 3D understanding of roofs containing imagery, digital surface models, and 3D roof geometries captured from over 20k residential inspections.
WACV 2024
Revisiting pre-trained remote sensing model benchmarks: resizing and normalization matters We perform a fair and large-scale evaluation of remote sensing foundation models and find chip/tile size and normalization preprocessing to be paramount for achieving peak performance for each individual pretrained model.
ArXiv 2023
SSL4EO-L: Datasets and Foundation Models for Landsat Imagery We introduce SSL4EO-L, the first ever dataset designed for self-supervised learning for Earth Observation for the Landsat family of satellites.
NeurIPS 2023
Solar Panel Mapping via Oriented Object Detection
C. Wallace, I. Corley, J. Lwowski
We detail Zeitview's deployed end-to-end deep learning framework for detecting solar panels using rotated object detection architectures in large scale solar farms.
ICLR 2023 Tackling Climate Change with Machine Learning Workshop
TorchGeo: Deep Learning with Geospatial Data 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
I. Corley, P. Najafirad
We propose Contrastive Surface-Image Pretraining (CSIP) for joint learning a latent space which extracts surface level features from optical RGB imagery which we show improves the performance in surface relevant tasks e.g. building segmentation and change detection.
ICIP 2022
Self-Supervised Representation Learning Enhances Broad Area Search in Multi-Temporal Satellite Imagery We propose pretraining on satellite image time-series (SITS) using the Contrastive Predictive Coding (CPC) self-supervised learning method which dramatically improves the performance of Broad Area Search (BAS) tasks in limited labeled data settings.
Destruction of Image Steganography using Generative Adversarial Networks
I. Corley, J. Lwowski, J. Hoffman
We propose a Generative Adversarial Network (GAN) based method we coin Deep Digital Steganography Purifier (DDSP), which is optimized to remove steganographic content embedded in image pixels without compromising the perceptual quality of the original image.
ArXiv 2019
DomainGAN: Generating Adversarial Examples to Attack Domain Generation Algorithm Classifiers
I. Corley, J. Lwowski, J. Hoffman
We propose DomainGAN which generates adversarial domains with near identical characteristics to benign domains and which greatly evades SOTA DGA classifiers.
ArXiv 2019
Deep EEG Super-Resolution: Upsampling EEG Spatial Resolution with Generative Adversarial Networks
I. Corley, Y. Huang
We train a Generative Adversarial Network (GAN) to upsample the spatial resolution of EEG datasets by generating realistic signals, eliminating the need for expensive EEG hardware.
IEEE Biomedical & Health Informatics (BHI) 2019