Feature Dimensionality Reduction for Video Affect Classification: A Comparative Study

Abstract

Affective computing has become a very important research area in human-machine interaction. However, affects are subjective, subtle, and uncertain. So, it is very difficult to obtain a large number of labeled training samples, compared with the number of possible features we could extract. Thus, dimensionality reduction is critical in affective computing. This paper presents our preliminary study on dimensionality reduction for affect classification. Five popular dimensionality reduction approaches are introduced and compared. Experiments on the DEAP dataset showed that no approach can universally outperform others, and performing classification using the raw features directly may not always be a bad choice.

Pool-Based Sequential Active Learning for Regression

“…It consists of 167 acoustic emotional stimuli for experimental investigations of emotion and attention. 76 acoustic features were extracted [17], and principle component analysis was used to reduce them to 10 features. The goal was to estimate the continuous arousal value from these 10 features.…”

OMG - Emotion Challenge Solution

“…We then computed the mean and/or variance of these frame-level features, resulting in a total of 76 audio features, as shown in Table I. These 76 features have been used in our previous research [4]. In this solution, instead of using these 76 features directly, we first clipped each feature into its [2,98] percentile interval (e.g., all values smaller than 2 percentile were replaced by the value at 2 percentile, and all values larger than 98 percentile were replaced by the value at 98 percentile), normalized to [0, 1], and then used RReliefF [7] to sort the features according to their importance.…”