- I am giving an invited talk at the Video Summarization for Large-scale Analytics Workshop @ ICCV 2015.
- DOE TERRA Project starting active sensing and prediction for high throughput crop phenotyping (collaborating with Roman Garnett).
- 2015 ICCV paper accepted: "Building Dynamic Cloud Maps from the Ground Up". This paper answers the question: "How could we use 4388 webcams distributed all over the planet as a replacement for satellites to create continental scale cloud maps".
- 2015 3DV paper accepted: "The Geometry of Colorful, Lenticular Fiducial Markers". This paper answers the question: How can we use lenticular arrays to create fiducial markers whose color varies with relative pose, and how could we use such markers to make pose estimation faster and more accurate?
- I am teaching Advanced Algorithms, CSE 441/451 this fall.
- We presented our ICCP 2015 paper "Structured Light Field Design for Correspondence Free Rotation Estimation"
- We presented our WACV 2015 paper, Characterizing Feature Matching Performance over Long Time Periods, exploring how and why feature matching degrades over years.
ResearchI am a professor of Computer Science and Engineering at Washington University in St. Louis. My research focuses on understanding images and how they change by exploring the geometry of the light field --- the 3-D pattern of light rays that characterize our visual environment. The light field is a vehicle to explore technical challenges including how to:
- design materials or objects that are extremely easy to measure and quantify,
- design cameras that are optimized for tasks involving measurement and manipulation, and
- measure the environment and activities in outdoor locations in the context of very long term, uncontrolled image data sets.
- democratize the ability of all people to create, understand and contribute to citizen science and citizen journalism projects
- improve the ability to measure the effects of changing climate on our eco-system, by visualy documenting and measuring phenological changes at the continental scale, working with Andrew Richardson.
- characterize how people use public spaces, and how that responds to physical activity policy and built environment change", with the incomparable Aaron Hipp (funded through an NSF R21 grant).
- develop fundamentally new representations of video variations over time, with Bill Freeman, Wojciech Matusik, Hanspeter Pfister, and Noah Snavely.
- Finding the lost grave of a Jane Doe crime victim in a 30 year old case from old imagery (Fox News story).
- CACM blurb:
cloudly day time-lapse to solve for 3D scene structure.
- New Scientist blurb: integrate webcams into Google Earth.
- The Engineering School Magazine recently wrote up a non-technical version of the research in my lab.
Dataset, Apps, and Frequenty Requested Code
- The Archive of Many Outdoor Scenes (AMOS).
- Citizen Science Repeat Photography App and Data (rePhoto).
- Code (12 KB zip file) for matlab and web visualizations of image manifolds, from the paper: "A Survey of Manifold Learning for Images", Robert Pless and Richard Souvenir, IPSJ Transactions on Computer Vision and Applications, vol. 1 pp. 83-94, 2009. (bib)
- Code (2.3 MB zip file) from our paper: "Extrinsic Calibration of a Camera and Laser Range Finder", Qilong Zhang, Robert Pless, IROS 2004, (bib),