About Me
I am currently a Postdoctoral Fellow in the Daniel Guggenheim School of Aerospace Engineering at Georgia Institute of Technology. I received my M.S. and Ph.D. from the Robotics Institute at the University of Michigan where I was a member of the Deep Robot Optical Perception Laboratory. I hold a B.S.E. in Mechanical and Aerospace Engineering with a Certificate in Applications of Computing from Princeton University. I have previously held appointments at Woods Hole Oceanographic Institution and the Australian Centre for Field Robotics.
My research interests span robotics, computer vision, and machine learning with focus on enabling autonomy in dynamic, unstructured, or remote environments across field robotics applications. My dissertation research focused on machine learning for robotic perception in underwater environments. I have also collaborated with the Ford Center for Autonomous Vehicles to improve perception for autonomous vehicles in urban environments.
Recent News
Check out The Future of Unmanned Ocean Innovation & Exploration - World Oceans Day Panel! I was really excited to join in this discussion hosted by Seaworthy Collective and am looking forward to their future mentorship and networking events focused on fostering innovation in ocean science and engineering.
I gave an IRIM Robotics Seminar at Georgia Tech.
I started as a Postdoctoral Fellow in Aerospace Engineering at Georgia Tech.
I was invited to give a keynote talk at the Underwater Robotics Perception Workshop at ICRA '19! [Slides]
I will be starting as an Assistant Professor in Naval Architecture and Marine Engineering and Robotics at the University of Michigan in 2021!
Research
Unsupervised Learning for Underwater Imagery
Deep learning has demonstrated great success in modeling complex nonlinear systems but requires a large amount of training data, which is difficult to compile in subsea environments. My dissertation work leverages physics-based models of underwater image formation to develop unsupervised learning approaches to advance perceptual capabilities of underwater robots. In particular, I have focused on unsupervised learning for color correction and depth estimation of monocular and stereo underwater imagery.
Perception for Autonomous Driving
I have also collaborated with the Ford Center for Autonomous Vehicles at University of Michigan to improve perception for autonomous vehicles in urban environments. The videos below show results from our work on transferring sensor-based effects from real data to simulated data to improve results of training on simulated data for the task of object detection.
Light Field Imaging in Underwater Environments
Light field cameras have a microlens array between the camera's main lens and image sensor, enabling recovery of a depth map and high resolution image from a single optical sensor. Our research has focused on leveraging light field cameras to improve underwater perception, with tasks including real-time 3D reconstruction and underwater image dehazing.
Underwater Bundle Adjustment
Our work developing underwater bundle adjustment integrates color correction into the structure recovery procedure for multi-view stereo reconstruction in underwater environments.
Robotic Survey of Sunken Pirate City
Our team conducted a robotic survey of the submerged city of Port Royal, Jamaica to create a 3D reconstruction of the marine archaeological site. [Read more]
Publications
“Sensor Transfer: Learning Optimal Sensor Effect Image Augmentation for Sim-to-Real Domain Adaptation.” (Alexandra Carlson, Katherine A. Skinner, Ram Vasudevan and Matthew Johnson-Roberson), In IEEE Robotics and Automation Letters (RA-L), 2019.
“DispSegNet: Leveraging Semantics for End-to-End Learning of Disparity Estimation from Stereo Imagery.” (Junming Zhang, Katherine A. Skinner, Ram Vasudevan and Matthew Johnson-Roberson), In IEEE Robotics and Automation Letters (RA-L), 2019.
“UWStereoNet: Unsupervised Learning for Depth Estimation and Color Correction of Underwater Stereo Imagery.” (Katherine A. Skinner, Junming Zhang, Elizabeth Olson and Matthew Johnson-Roberson), In IEEE International Conference on Robotics and Automation (ICRA), Montreal, Canada, 2019.
“Synthetic Data Generation for Deep Learning of Underwater Disparity Estimation.” (Elizabeth Olson, Corina Barbalata, Katherine A. Skinner and Matthew Johnson-Roberson), In Proceedings of the IEEE/MTS OCEANS Conference and Exhibition, Charleston, USA, October 2018.
“Modeling Camera Effects to Improve Visual Learning from Synthetic Data.” (Alexandra Carlson*, Katherine A. Skinner*, Ram Vasudevan and Matthew Johnson-Roberson), In Proceedings of the European Conference on Computer Vision Workshop on Visual Learning and Embodied Agents in Simulated Environments (ECCV Workshops), Munich, Germany, 2018. *The authors contributed equally to this work.
"WaterGAN: Unsupervised Generative Network to Enable Real-time Color Correction of Monocular Underwater Images." (Jie Li*, Katherine A. Skinner*, Ryan Eustice, and Matthew Johnson-Roberson), In IEEE Robotics and Automation - Letters, 2017. *The authors contributed equally to this work. [BibTeX, PDF]
”Multi-view 3D Reconstruction in Underwater Environments: Evaluation and Benchmark.” (Eduardo Iscar, Katherine A. Skinner, and Matthew Johnson-Roberson), In MTS/IEEE OCEANS, 2017.[BibTeX]
”Underwater Image Dehazing with a Light Field Camera.” (Katherine A. Skinner and Matthew Johnson-Roberson), In IEEE Conference on Computer Vision and Pattern Recognition – Workshops, 2017 (Oral Presentation).[BibTeX]
"Automatic Color Correction for 3D Reconstruction of Underwater Scenes ." (Katherine A. Skinner, Eduardo Iscar, and Matthew Johnson-Roberson), In IEEE International Conference on Robotics and Automation, 2017. [BibTeX]
"Towards Real-time Underwater 3D Reconstruction with Plenoptic Cameras." (Katherine A. Skinner and Matthew Johnson-Roberson), In IEEE/RSJ International Conference on Intelligent Robots and Systems, 2016. [BibTeX][PDF]
"Detection and Segmentation of Underwater Archaeological Sites Surveyed with Stereo-Vision Platforms." (Katherine A. Skinner and Matthew Johnson-Roberson), In MTS/IEEE OCEANS, 2015. [BibTeX][PDF]
"Bathymetric Factor Graph SLAM with Sparse Point Cloud Alignment." (Vittorio Bichucher, Jeffrey M. Walls, Paul Ozog, Katherine A. Skinner, and Ryan M. Eustice), In MTS/IEEE OCEANS, 2015.[BibTeX, PDF]
Teaching
AE 2220-B: Dynamics, Georgia Institute of Technology (Spring 2020) - Instructor
An introduction to kinematics and kinetics of rigid bodies in both plane and 3-D motion.
NAVARCH 599: Underwater Robotics, University of Michigan (Winter 2018) - Guest Lecturer
A special topics course on underwater robotics offered to upper class undergraduates and graduate students.
EECS 442: Computer Vision, University of Michigan (Fall 2016) - Graduate Student Instructor
An introduction to 2D and 3D computer vision offered to upper class undergraduates and graduate students. Topics include camera models, multi-view geometry, stereo reconstruction, low-level image processing methods, segmentation, clustering, and high-level vision techniques such as object recognition.
ENG 100: Introduction to Engineering, University of Michigan (Fall 2016) - Guest Lecturer
An introduction to underwater vehicle design for freshmen undergraduates. [Lecture Slides]
Outreach
GO-GIRL
Our lab coordinated with Wayne State University's Gaining Options - Girls Investigate Real Life (GO-GIRL) program to design a summer workshop for high school girls as an introduction to engineering. The workshop involved building a SeaPerch remotely operated vehicle (ROV) and testing it in a lab setting and in a local pond to gather data. [Tutorial]
Discover Engineering
The Robotics Graduate Student Council organized a workshop for UM College of Engineering's Discover Engineering program. The program is geared towards 8th-10th graders who are children of UM alumni. Our workshop featured a fun robot bowling challenge! [Workshop Slides]
Robotics Day
UM hosts Robotics Day each year, bringing together industry, academia, and government agencies to highlight robotics advances across Michigan. This year the Robotics Graduate Student Council hosted a table where young roboticists could design, build, and test their very own drawing robots. [Workshop Handout]
More Involvement
Robotics Graduate Student Council - Co-founder (Past: President, Treasurer, Outreach Chair)
NSF IS-GEO Research Coordination Network - Early Career Committee Member
Undergraduate Research Opportunity Program - Mentor
Mentor2Youth Program - Instructor [Tutorial]
GradSWE Elementary School Science Program - Instructor
Robotics Day Planning Committee - Student Representative
A World in Motion - Instructor