Tangent bundle for human action recognition
Yui Man Lui, J. Ross Beveridge
Abstract
Yui Man Lui, J. Ross Beveridge
Abstract
Common human actions are instantly recognizable by people and increasingly machines need to understand this language if they are to engage smoothly with people. Here we introduce a new method for automated human action recognition. The proposed method represents videos as a tangent bundle on a Grassmann manifold. Videos are expressed as third order tensors and factorized to a set of tangent spaces. Tangent vectors are then computed between elements on a Grassmann manifold and exploited for action classification. In particular, logarithmic mapping is applied to map a point from the manifold to tangent vectors centered at a given element. The canonical metric is used to induce the intrinsic distance for a set of tangent spaces. Empirical results show that our method is effective on both uniform and non-uniform backgrounds for action classification. We achieve recognition rates of 91% on the Cambridge gesture dataset, 88% on the UCF sport dataset, and 97% on the KTH human action dataset. Additionally, our method does not require prior training.
OpenAlex reports 48 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Common human actions are instantly recognizable by people and increasingly machines need to understand this language if they are to engage smoothly with people. Here we introduce a new method for automated human action recognition. The proposed method represents videos as a tangent bundle on a Grassmann manifold. Videos are expressed as third order tensors and factorized to a set of tangent spaces. Tangent vectors are then computed between elements on a Grassmann manifold and exploited for action classification. In particular, logarithmic mapping is applied to map a point from the manifold to tangent vectors centered at a given element. The canonical metric is used to induce the intrinsic distance for a set of tangent spaces. Empirical results show that our method is effective on both uniform and non-uniform backgrounds for action classification. We achieve recognition rates of 91% on the Cambridge gesture dataset, 88% on the UCF sport dataset, and 97% on the KTH human action dataset. Additionally, our method does not require prior training.
Key concepts: Tangent bundle, Tangent vector, Tangent space, Action (physics), Manifold (fluid mechanics), Set (abstract data type), Metric (unit), Mathematics