Faculty and students are exploring a number of critical problems in the area of computer vision, with a focus on the analysis and modeling of visual scenes from static images as well as video sequences. Research goals include: i) the semantic understanding of materials, objects, and actions within a scene; ii) modeling the spatial organization and layout of the scene and its behavior in time. The algorithms developed in this area of research enable the design of machines that can perform real-world visual tasks such as autonomous navigation, visual surveillance, or content-based image and video indexing.
Voxel51, a U-M startup led by Prof. Jason Corso, uses custom AI to continuously track vehicle, cyclist, and pedestrian traffic in real time at some of the most visited places in the world.
Prof. Hun-Seok Kim helped design iGYM, an augmented reality system that allows disabled and able-bodied people to play physical games together.
The frame in which a human marks out the boundaries of an object makes a huge difference in how well AI software can identify that object through the rest of the video.
The most exciting use of AI for me focuses around a better collective use of our available resources, says Corso.
PhD student Jean Young Song offers an improved solution to the problem of image segmentation.
Kyle Min researches how computer vision can analyze law enforcement body cameras.
DARPA is trying to build a system that can turn large data sets into models that can make predictions, and U-M is in on the project.
Centralizing available data in the intelligent systems community through a COmputer Vision Exchange for Data, Annotations and Tools, called COVE.
Researchers have found a way to improve a computer’s human-tracking accuracy by looking at where the targets are going, but also at what they’re doing.
“We have pioneered an integrated scene understanding framework that enables the automatic tracking of structural changes, allowing data to be collected easily.”
Computer Vision seeks to imitate humans’ ability to recognize objects, navigate scenes, reconstruct layouts, and understand the geometric space and semantic meaning.
Srinath’s research focuses on using computer vision techniques such as markerless camera tracking for creating augmented reality AR environments.
Bao’s research is in Semantic Structure from Motion, a new framework for jointly recognizing objects as well as reconstructing their underlying 3D geometry.
“This book organizes and introduces major concepts in 3D scene and object representation and inference from still images.”
With D4AR models, you can monitor progress, productivity, safety, quality, constructability and even site logistics remotely.