Spatio-temporal Co-Occurrence Characterizations for Human Action Classification

TL;DR

This paper introduces a novel approach for human action classification using spatio-temporal co-occurrence features, significantly improving accuracy over traditional bag-of-video-words methods on standard datasets.

Contribution

The paper proposes a new co-occurrence based feature extraction method and demonstrates its effectiveness in enhancing human action classification accuracy.

Findings

Achieved state-of-the-art results on KTH and UCF-Sports datasets.

Demonstrated improved classification performance using co-occurrence features.

Validated the effectiveness of the proposed method over standard approaches.

Abstract

The human action classification task is a widely researched topic and is still an open problem. Many state-of-the-arts approaches involve the usage of bag-of-video-words with spatio-temporal local features to construct characterizations for human actions. In order to improve beyond this standard approach, we investigate the usage of co-occurrences between local features. We propose the usage of co-occurrences information to characterize human actions. A trade-off factor is used to define an optimal trade-off between vocabulary size and classification rate. Next, a spatio-temporal co-occurrence technique is applied to extract co-occurrence information between labeled local features. Novel characterizations for human actions are then constructed. These include a vector quantized correlogram-elements vector, a highly discriminative PCA (Principal Components Analysis) co-occurrence vector and a Haralick texture vector. Multi-channel kernel SVM (support vector machine) is utilized for classification. For evaluation, the well known KTH as well as the challenging UCF-Sports action datasets are used. We obtained state-of-the-arts classification performance. We also demonstrated that we are able to fully utilize co-occurrence information, and improve the standard bag-of-video-words approach.