 |
    |
Continuous Conditional Neural Fields for Structured Regression*
Tadas Baltrušaitis1, Peter Robinson1, and Louis-Philippe Morency2
1Computer Laboratory, University of Cambridge, UK
tadas.baltrusaitis@cl.cam.ac.uk
peter.robinson@cl.cam.ac.uk
2Institute for Creative Technologies, University of Southern California, CA, USA
morency@ict.usc.edu
Abstract. An increasing number of computer vision and pattern recognition problems require structured regression techniques. Problems like human pose estimation, unsegmented action recognition, emotion prediction and facial landmark detection have temporal or spatial output dependencies that regular regression techniques do not capture. In this paper we present continuous conditional neural fields (CCNF) – a novel structured regression model that can learn non-linear input-output dependencies, and model temporal and spatial output relationships of varying length sequences. We propose two instances of our CCNF framework: Chain-CCNF for time series modelling, and Grid-CCNF for spatial relationship modelling. We evaluate our model on five public datasets spanning three different regression problems: facial landmark detection in the wild, emotion prediction in music and facial action unit recognition. Our CCNF model demonstrates state-of-the-art performance on all of the datasets used.
Keywords: Structured regression, Landmark detection, Face tracking
Electronic Supplementary Material:
LNCS 8692, p. 593 ff.
lncs@springer.com
© Springer International Publishing Switzerland 2014