2012 Poster Sessions : Object Class Detectors for 3D Scene Understanding

Student Name : Michael Stark
Advisor : None
Research Areas: Artificial Intelligence
Current object class recognition systems typically target 2D bounding box localization, encouraged by benchmark data sets, such as Pascal VOC. While this seems suitable for the detection of individual objects, higher-level applications such as 3D scene understanding or 3D object tracking would benefit from more fine-grained object hypotheses incorporating 3D geometric information, such as viewpoints or the locations of individual parts.

In this work, we help narrowing the representational gap between the ideal input of a scene understanding system and object class detector output, by designing detectors particularly tailored towards 3D geometric reasoning. In particular, we present three different variants of object class detectors that provide both estimates of object viewpoint and 3D parts that are consistent across viewpoints. We experimentally verify that adding 3D geometric information comes at minimal performance loss w.r.t. 2D bounding box localization, but outperforms prior work in 3D viewpoint estimation and ultra-wide baseline matching.

Michael Stark is a postdoc with the Max Planck Center for Visual Computing and Communication at Stanford, advised by Daphne Koller. He obtained his Ph.D. (Dr.-Ing.) from TU-Darmstadt, Germany, in 2010 under the supervision of Bernt Schiele. He was a postdoc with the Max Planck Institute for Informatics, Saarbruecken, Germany from 2010 to 2011.