Jinwei Ye
George Mason University, VA·Computer Science·Early-career
Cameras and computer vision algorithms usually work with only the color and brightness of light. They struggle when light is bent, scattered or distorted on its way to the camera, as happens underwater, through fog, or across long stretches of turbulent air.
- Fields
Computer Science
Engineering
- Topics
computer vision
computational imaging
computational photography
- Grants
- 3 known grants, through Jun 2028
What they work on
- Model how the polarization of light changes when it reflects off complex surfaces, using physics-based layered material models and a shared high-precision polarimetric reflectance dataset.
- Develop a voxel-based model of polarized light traveling through volumes and murky media such as fog and water, and use analysis by synthesis to recover depth, shape, reflectance and object identity.
- Separate object motion from turbulence-induced motion in long-range images from drones and distant cameras, using neural motion-field estimators based on physical velocity fields.
- Estimate turbulence strength and flow in air and water from ordinary video, and use these motion fields to guide intelligent foveation for long-range vision tasks.
- Build an angularly sampled imaging system that emits and collects light rays at chosen angles, to analyze how light bends, scatters and is absorbed underwater.
- Develop underwater reflectance models and 3D shape reconstruction algorithms based on underwater ray optics, and integrate them into vision-based tracking for underwater vehicles.
More about the research
The lab's projects model the physics of how light travels and interacts with surfaces and media. They then use those models, together with new camera hardware and algorithms, to let machines see shape, depth and materials in conditions where ordinary cameras fail.
Why it matters
These methods could give robots, drones and underwater vehicles reliable vision where cameras now fail: in fog, haze, turbulent air and murky water. Possible beneficiaries include autonomous navigation, smart manufacturing, drone tracking and long-range monitoring. They also include seafloor mapping for marine research and the oil and gas industry, and imaging through tissue. The work may lower the cost and risk of ocean exploration.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Nianyi Li
Clemson University, SC·Computer Science·Early-career
Build a multi-sensor imaging system that combines RGB cameras, depth sensors, LiDAR and event cameras to capture diverse data from unstructured outdoor environments.
Noah Snavely
Cornell University, NY·Computer Science
Develop compositional implicit neural scene representations that pair a network's ability to capture fine detail with physical models of scene structure, such as light fields.
Robert Guy
University of California-Davis, CA·Mathematics & Statistics
Measure flow patterns and forces on dynamically similar remotely operated vehicles to test how metachronal rowing, a back-to-front wave of paddle strokes, generates thrust and lift across paddle Reynolds numbers from 0.01 to 10,000.
The school
George Mason University: 15 labs on PI Pages, $282M a year on research, 297 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.