Research >> Dynamic scenes and 3D change detection

 

Dynamic scenes and 3D change detection

 

Building a model of a scene from a series of 3D scans taken from
different viewpoints and at different times is a difficult problem. A
major challenge is that objects in the scene may move (or be moved)
between different scans. Points in the scan from these objects should
be highlighted or removed, and the final reconstruction should only
contain the static components of the scene. Our approach is a hybrid
of surface and volumetric representations, and we have recently made
substantial progress on a new representation based on point-set
surfaces.

 

Faculty:

Chuck Stewart

Daniel Freedman

Rich Radke
Students:

Eric Smith

Brad King

 

 



Home | People | Research | Publications | Contact | News/Events | Download | Links