- Oct 2024: Excited to announce I have successfully defended my Ph.D. Dissertation entitled Multimodal Learning for Infrastructure Mapping in Remote Sensing!
- Oct 2024: Our paper Barely-Visible Surface Crack Detection for Wind Turbine Sustainability has been accepted to IROS 2024 and won Best Application Paper Runner-Up!
- Sep 2024: Our paper Depth Any Canopy: Leveraging Depth Foundation Models for Canopy Height Estimation has been accepted to the ECCV 2024 Workshop on Computer Vision for Ecology (CV4E).
- Jul 2024: Our paper Estimating Earthquake Magnitude in Sentinel-1 Imagery via Ranking has been accepted to the ECML-PKDD 2024 Workshop on Machine Learning for Earth Observation (MACLEAN).
- May 2024: Excited to announce I will be joining Microsoft Research’s AI For Good Lab as a Ph.D. Research Intern for the summer! I will be researching Multimodal Pretraining methods for Large-Scale Geospatial Vision-Language Datasets.
- Apr 2024: Our paper Revisiting pre-trained remote sensing model benchmarks: resizing and normalization matters has been accepted to the CVPR 2024 Perception Beyond the Visible Spectrum (PBVS) Workshop.
- Mar 2024: Our paper Seeing the roads through the trees: A benchmark for modeling spatial dependencies with aerial imagery has been accepted to IEEE IGARSS 2024.
- Mar 2024: Our paper A Change Detection Reality Check has been accepted to the ICLR 2024 Machine Learning for Remote Sensing (ML4RS) Workshop.
- Feb 2024: We present A Change Detection Reality Check, which provides 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.
- Jan 2024: We present Seeing the roads through the trees: A benchmark for modeling spatial dependencies with aerial imagery, which introduces a novel remote sensing dataset and evaluates a model’s ability to learn global and long-range spatial context by extracting roads in aerial imagery containing large gaps due to occlusion by tree canopy.
- Oct 2023: Our paper ZRG: A Dataset for Multimodal 3D Residential Rooftop Understanding has been accepted to WACV 2024.
- Sep 2023: Our paper SSL4EO-L: Datasets and Foundation Models for Landsat Imagery has been accepted to NeurIPS 2023.
- May 2023: We present Revisiting pre-trained remote sensing model benchmarks: resizing and normalization matters, which performs 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.
- Apr 2023: We present Single-View Height Estimation with Conditional Diffusion Probabilistic Models, which trains a Denoising Diffusion Probabilistic Model (DDPM) conditioned on aerial imagery to estimate pixelwise height from a monocular view.
- Jan 2023: Our paper Solar Panel Mapping via Oriented Object Detection has been accepted to the ICLR 2023 Workshop: Tackling Climate Change with Machine Learning.
- Sep 2022: Our paper Torchgeo: Deep Learning with Geospatial Data has been accepted to ACM SIGSPATIAL 2022.
- Jun 2022: Our paper Supervising Remote Sensing Change Detection Models with 3D Surface Semantics has been accepted to ICIP 2022.
- Apr 2022: Our paper Self-Supervised Representation Learning Enhances Broad Area Search in Multi-Temporal Satellite Imagery has been accepted to IGARSS 2022.