Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
3D Pose Estimation and Segmentation using Specular Cues Ju Yong Chang1,
 

Summary: 3D Pose Estimation and Segmentation using Specular Cues
Ju Yong Chang1,
Ramesh Raskar2
Amit Agrawal1
1
Mitsubishi Electric Research Labs (MERL), 201 Broadway, Cambridge, MA
2
MIT Media Lab, 20 Ames St., Cambridge, MA
juyong.chang@gmail.com
Abstract
We present a system for fast model-based segmentation
and 3D pose estimation of specular objects using appear-
ance based specular features. We use observed (a) specular
reflection and (b) specular flow as cues, which are matched
against similar cues generated from a CAD model of the
object in various poses. We avoid estimating 3D geome-
try or depths, which is difficult and unreliable for specu-
lar scenes. In the first method, the environment map of the
scene is utilized to generate a database containing synthe-
sized specular reflections of the object for densely sampled

  

Source: Agrawal, Amit - Mitsubishi Electric Research Labs

 

Collections: Engineering