Computer Vision News - March 2019

Method As mentioned, the Kalman Filter consists of two stages: the prediction stage and the update stage. The prediction stage computes an estimated scalar ෞ . The update stage (also known as the correction step), fuses this estimate ( ෞ ) with the latest measurements , where ∈ {1, . . . , } is the time and ∈ {1, . . . , } is the m-th classifier. The prediction stage: the new prediction ෞ is estimated based on the last prediction ෟ −1 , according to the formula ෞ = ሶ ⋅ −1 . The original Kalman Filter formula includes an update element u ; however, it is meaningless here, as it refers to a command or instruction given to a robot or autonomous vehicle, and the paper deals with a classifier. ෞ is the covariance of the prediction and is equal to ෞ = ⋅ ⋅ + : it is obtained by combining the a posteriori covariance with an additional covariance , which models the process noise: The update stage is performed for each classifier m and it consists of the following three sub-component computations: = − ℎ ⋅ ෞ = ℎ ⋅ ෞ ⋅ ℎ + = ෞ ⋅ ℎ ⋅ −1 where h is the observation model mapping the prediction to the new estimate, and is the observation noise. The outcomes of this stage are used to update the estimate and covariance of the next stage, based on the following formula: = ෝ + ⋅ = ෞ − ⋅ ⋅ When there are missing values, that is for some classifiers there is no decision -- the output of that classifier ( ) is set as 0.5 and the observation noise of that classifier is raised. The authors expanded the classifiers’ decision range to include the possibility of rejection, that is, classifiers may decide to return a no- classification output, rather than their estimate, based on a too-low confidence measure. Due to the temporal structure of the method (the fact that there are a number of classifications at every point in time), the missing classification results can be estimated based on the results of the other classifiers. The architecture proposed in the paper is as follow - at every point in time, each classifier m produces a classification decision and a corresponding confidence measure, which serves for making rejection decisions. Dataset The paper used the AVEC dataset, presented in 2011 as a benchmark for user Computer Vision News Fusion for Affective State Recognition 5 Research Computer Vision News