Nathan Silberman

I currently work at Google Research on computer vision related projects. I completed my PhD in 2015 and was advised by Rob Fergus and David Sontag. My research interests lie in the realm of scene understanding. Specifically, I am interested in representations of scenes that are richer than pure object labeling in the image plane.

Email: [my last name]@cs.nyu.edu


A Contour Completion Model for Augmenting Surface Reconstructions was accepted at ECCV 2014

Instance Segmentation of Indoor Scenes using a Coverage Loss was accepted at ECCV 2014

I'm helping to organize the Reconstruction Meets Recognition Challenge at ECCV 2014. The challange datasets are now available.

I'm helping to organize the Reconstruction Meets Recognition Challenge at ICCV 2013.

Code now available for Indoor Segmentation and Support Inference from RGBD Images.

Indoor Segmentation and Support Inference from RGBD Images accepted as an Oral at ECCV

NYU Depth V2 Released.