Home/Learning Message-Passing Inference Machines for Structured Prediction

Learning Message-Passing Inference Machines for Structured Prediction

Stephane Ross, Daniel Munoz, Martial Hebert and J. Andrew (Drew) Bagnell
Conference Paper, Carnegie Mellon University, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June, 2011

Download Publication (PDF)

Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author’s copyright. These works may not be reposted without the explicit permission of the copyright holder.


Nearly every structured prediction problem in computer vision requires approximate inference due to large and com- plex dependencies among output labels. While graphical models provide a clean separation between modeling and inference, learning these models with approximate infer- ence is not well understood. Furthermore, even if a good model is learned, predictions are often inaccurate due to approximations. In this work, instead of performing infer- ence over a graphical model, we instead consider the in- ference procedure as a composition of predictors. Specif- ically, we focus on message-passing algorithms, such as Belief Propagation, and show how they can be viewed as procedures that sequentially predict label distributions at each node over a graph. Given labeled graphs, we can then train the sequence of predictors to output the correct label- ings. The result no longer corresponds to a graphical model but simply defines an inference procedure, with strong the- oretical properties, that can be used to classify new graphs. We demonstrate the scalability and efficacy of our approach on 3D point cloud classification and 3D surface estimation from single images.

BibTeX Reference
title = {Learning Message-Passing Inference Machines for Structured Prediction},
author = {Stephane Ross and Daniel Munoz and Martial Hebert and J. Andrew (Drew) Bagnell},
booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
sponsor = {National Sciences and Engineering Research Council of Canada (NSERC), QinetiQ North America Robotics Fellowship},
school = {Robotics Institute , Carnegie Mellon University},
month = {June},
year = {2011},
address = {Pittsburgh, PA},