Dominant Codewords Selection with Topic Model for Action Recognition

TL;DR

This paper introduces a novel action recognition framework that leverages dominant codewords and topic modeling to improve the representation of human activities by focusing on motion primitives and reducing noise.

Contribution

It presents a new method combining LDA-based topic modeling with dominant codewords for more accurate human activity recognition.

Findings

Effective in recognizing actions across four datasets

Improves motion primitive representation by eliminating noise

Enhances activity classification accuracy

Abstract

In this paper, we propose a framework for recognizing human activities that uses only in-topic dominant codewords and a mixture of intertopic vectors. Latent Dirichlet allocation (LDA) is used to develop approximations of human motion primitives; these are mid-level representations, and they adaptively integrate dominant vectors when classifying human activities. In LDA topic modeling, action videos (documents) are represented by a bag-of-words (input from a dictionary), and these are based on improved dense trajectories. The output topics correspond to human motion primitives, such as finger moving or subtle leg motion. We eliminate the impurities, such as missed tracking or changing light conditions, in each motion primitive. The assembled vector of motion primitives is an improved representation of the action. We demonstrate our method on four different datasets.