Congrats to Hunter Blanton (hblanton.github.io) for successfully defending his Ph.D. thesis on improving blackbox camera pose estimation (e.g., PoseNet) by explicitly decomposing the internals to be geometry aware (arxiv.org/abs/2012.12360)!

Live workshop on #IWDSC2021 ICCV2021 begins at 8 am EDT (12 pm GMT) today. Join us! Follow our insightful speakers (Nathan Jacobs, Sara Hooker,Georgia Gkioxari,Juan Carlos Niebles) and our panellists presenting their innovative works! iwdsc.github.io/program/

Dynamic Feature Alignment for Semi-supervised Domain Adaptation British Machine Vision Conference (BMVC) why: we often have some labels in the target domain idea: align the target features to a set of dynamically updated class prototypes to avoid problems with class imbalances w/ Yu Zhang and GB. Tony Liang

I'm excited to announce I'll be joining the CS&E dept at Washington University in St. Louis in Fall 2022! It's been a great decade+ UK Computer Science (with a wonderfully supportive department). I'll be looking for new collaborators, postdocs, graduate students, housing, ... stay tuned!

📢📢📢 New paper published at AGU "Earth and Space Science" Journal Automatic Segmentation of Sinkholes Using a Convolutional Neural Network w/ Dr. Nathan Jacobs (Multimodal Vision Research Lab @ WashU) and Dr. J. Zhu from KY Geological Survey Paper: agupubs.onlinelibrary.wiley.com/doi/10.1029/20… Code: github.com/mvrl/sink-seg/

Our paper "Content-Based Detection of Temporal Metadata Manipulation" was accepted to the IEEE Trans. on Information Forensics and Security! Summary: Given an image and a geographic location, does this timestamp make sense? arxiv.org/abs/2103.04736 Lead author: Rafael Padilha

#CVPR2022 outcome: 1 paper accepted, 1 rejected. The review process is noisy but useful. Happy for the accepted paper and got some good pointers about the rejected one. Looking forward to #CVPR2025, hopefully in New Orleans 🙂 w/ Nathan Jacobs and Multimodal Vision Research Lab @ WashU

Congrats to Connor Greenwell (connorgreenwell.com) for successfully defending his PhD thesis in Computer Science UK Engineering! His research focused on image geo-localization with attention-based models. Next stop: Kitware!

Congrats to Yu Zhang (yuzhang03.github.io) for successfully defending his Ph.D. thesis in Computer Science UK Engineering! His research focused on unsupervised and semi-supervised domain adaptation across a range of computer vision tasks!

Looking for an exciting postdoc position? Help us (Michael D. Frachetti David B. Carter) combine satellite imagery and genomic + geospatial data to understand the interplay between natural environments, human mobility, and political borders. Apply here: apply.interfolio.com/124133

📢Joining Sara Beery Sara Beery and Diane Larlus Diane Larlus in delivering keynotes at #EARTHVISION2023 as part of upcoming #CVPR2025! #EarthVision aim is to foster collaboration between the computer vision and Earth Observation communities. EarthVision grss-ieee.org/events/earthvi…

Thanks to lab alumni M. Usman Rafique for sending us a care package of snacks for the #CVPR2025 deadline push!

Thanks EarthVision for the recognition, especially after a couple of rounds of rejections at other venues! Congrats to the team: Aayush (sites.wustl.edu/aayush/), Adeel Ahmad, Subash (subash-khanal.github.io), Srikumar (sites.wustl.edu/srikumarsastry/) and Hannah Kerner!

I'm so excited to finally introduce Fields of The World (FTW!) - the largest & most diverse ML dataset for field instance segmentation in satellite images. FTW has >70k samples from 24 countries (4 continents), with semantic & instance labels + Sentinel-2 RGB-NIR from 2 dates. 🧵