Belief Propagation Reloaded: Learning BP-Layers for Labeling Problems

Patrick Knöbelreiter*, Christian Sormann, Alexander Shekhovtsov, Friedrich Fraundorfer, Thomas Pock

*Korrespondierende/r Autor/-in für diese Arbeit

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandBegutachtung


It has been proposed by many researchers that combining deep neural networks with graphical models can create more efficient and better regularized composite models. The main difficulties in implementing this in practice are associated with a discrepancy in suitable learning objectives as well as with the necessity of approximations for the inference. In this work we take one of the simplest inference methods, a truncated max-product Belief Propagation, and add what is necessary to make it a proper component of a deep learning model: connect it to learning formulations with losses on marginals and compute the backprop operation. This BP-Layer can be used as the final or an intermediate block in convolutional neural networks (CNNs), allowing us to design a hierarchical model composing BP inference and CNNs at different scale levels. The model is applicable to a range of dense prediction problems, is well-Trainable and provides parameter-efficient and robust solutions in stereo, flow and semantic segmentation.

TitelProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2020
Herausgeber (Verlag)IEEEXplore
PublikationsstatusVeröffentlicht - 14 Juni 2020
Veranstaltung2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition: CVPR 2020 - virtuell, Virtual, USA / Vereinigte Staaten
Dauer: 14 Juni 202019 Juni 2020


Konferenz2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition
KurztitelCVPR 2020
Land/GebietUSA / Vereinigte Staaten

ASJC Scopus subject areas

  • Software
  • Maschinelles Sehen und Mustererkennung


Untersuchen Sie die Forschungsthemen von „Belief Propagation Reloaded: Learning BP-Layers for Labeling Problems“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren