The Community for Technology Leaders
Green Image
Issue No. 07 - July (2014 vol. 36)
ISSN: 0162-8828
pp: 1299-1311
Mihalis A. Nicolaou , Department of Computing, Imperial College London, London, SW, U.K.
Vladimir Pavlovic , Department of Computer Science, Rutgers University, Piscataway, NJ, USA
Maja Pantic , Department of Computing, Imperial College London, London, SW, U.K.
Fusing multiple continuous expert annotations is a crucial problem in machine learning and computer vision, particularly when dealing with uncertain and subjective tasks related to affective behavior. Inspired by the concept of inferring shared and individual latent spaces in Probabilistic Canonical Correlation Analysis (PCCA), we propose a novel, generative model that discovers temporal dependencies on the shared/individual spaces (Dynamic Probabilistic CCA, DPCCA). In order to accommodate for temporal lags, which are prominent amongst continuous annotations, we further introduce a latent warping process, leading to the DPCCA with Time Warpings (DPCTW) model. Finally, we propose two supervised variants of DPCCA/DPCTW which incorporate inputs (i.e., visual or audio features), both in a generative (SG-DPCCA) and discriminative manner (SD-DPCCA). We show that the resulting family of models (i) can be used as a unifying framework for solving the problems of temporal alignment and fusion of multiple annotations in time, (ii) can automatically rank and filter annotations based on latent posteriors or other model statistics, and (iii) that by incorporating dynamics, modeling annotation-specific biases, noise estimation, time warping and supervision, DPCTW outperforms state-of-the-art methods for both the aggregation of multiple, yet imperfect expert annotations as well as the alignment of affective behavior.
probability, behavioural sciences computing, computer vision, correlation methods, image fusion, learning (artificial intelligence)

M. A. Nicolaou, V. Pavlovic and M. Pantic, "Dynamic Probabilistic CCA for Analysis of Affective Behavior and Fusion of Continuous Annotations," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 36, no. 7, pp. 1299-1311, 2014.
285 ms
(Ver 3.3 (11022016))