Automatic analysis of human behaviour in large collections of videos is rapidly gaining interest, even more so with the advent of file sharing sites such as YouTube. From one perspective, it can be observed that the size of feature vectors used for human action recognition from videos has been increasing enormously in the last five years, in the order of ∼ 100-500K. One possible reason might be the growing number of action classes/videos and hence the requirement of discriminating features (that usually end up higher-dimensional for larger databases). In this paper, we review and investigate feature projection to reduce the dimensions of the high-dimensional feature vectors and show their effectiveness in terms of performance. We hypothesize that dimensionality reduction techniques often unearth latent structures in the feature space and are effective in applications such as fusion of high-dimensional features of different types; action recognition in untrimmed videos. We conduct all our experiments using a Bag-of-Words framework for consistency and results are presented on large class benchmark databases such as the HMDB51 and UCF101 datasets.