To really understand images and video, the computer must have some sense of the physical space of the depicted scenes. But how should we represent this physical space, and how can we recover it from imagery? Our focus is on creating simple but widely applicable models of scene geometry that can be recovered from a single image.
Beyond Line of Sight: code for ECCV'12 paper on labeling occluded background surfaces
Indoor Scene Dataset: includes indoor images labeled with box geometry, surface labels, and some objects
Geometric Context: includes images labeled according to surfaces and boundaries and region-based classification code
Automatic Photo Pop-up: includes dataset, code, and results
Beyond the line of sight: labeling the underlying surfaces
Ruiqi Guo and Derek Hoiem
ECCV, 2012.
[pdf]
[project]
Indoor Segmentation and Support Inference from RGBD Images
Nathan Silberman, Derek Hoiem, Pushmeet Kohli, and Rob Fergus
ECCV, 2012.
[pdf]
[project]
[slides]
Recovering Free Space of Indoor Scenes from a Single Image
Varsha Hedau and Derek Hoiem and David Forsyth
CVPR, 2012.
Recovering Occlusion Boundaries from an Image
D. Hoiem, A.A. Efros, and M. Hebert
IJCV, in press.
[pdf]
The final publication is available at www.springerlinks.com.
Recovering the Spatial Layout of Cluttered Rooms
V. Hedau, D. Hoiem, and D.A. Forsyth
ICCV 2009.
[pdf]
[project]
Closing the Loop on Scene Interpretation
D. Hoiem, A.A. Efros, and M. Hebert
CVPR 2008.
[pdf]
Seeing the World Behind the Image: Spatial Layout for 3D Scene Understanding
D. Hoiem
Doctoral Dissertation, CMU-RI-TR-07-28, Robotics Institute, Carnegie Mellon University, August 2007.
[pdf]
CMU School of Computer
Science Distinguished Dissertation Award
ACM Doctoral Dissertation Honorable
Mention
Learning to Extract Object Boundaries using Motion Cues
A.N. Stein, D. Hoiem, and M. Hebert
ICCV 2007.
[pdf]
Recovering Occlusion Boundaries from a Single Image
D. Hoiem, A.N. Stein, A.A. Efros, and M. Hebert
ICCV 2007.
[pdf]
Recovering Surface Layout from an Image
D. Hoiem, A.A. Efros, and M. Hebert
IJCV, Vol. 75, No. 1, October 2007.
[pdf]
The final publication is available at www.springerlinks.com.
Geometric Context from a Single Image
D. Hoiem, A.A. Efros, and M. Hebert
ICCV 2005. [pdf]
[project]
Automatic Photo Pop-up
D. Hoiem, A.A. Efros, and M. Hebert
ACM SIGGRAPH 2005. [pdf] [project]
NSF Award 0916014 for "Exploiting Geometric and Illumination Context in Indoor Scenes" (2009-2012)
NSF Award 0904209 for "Physically Grounded Object Recognition" (2009-2013)
Gift from Microsoft