Login Form

Editors

Activity recognition results on UCF Sports and Holywood2

Table above shows the results, obtained on UCF Sports dataset (http://crcv.ucf.edu/data/UCF_Sports_Action.php). We report recognition rate with respect to the number...


Read More...

Computational efficiency and parallel implementation

The developed algorithms are computationally effective and the compositional processing pipeline is well-suited for implementation on massively parallel architectures. Many...


Read More...

Motion hierarchy structure

Our model is comprised of three processing stages, as shown in the Figure. The task of the lowest stage (layers...


Read More...

Server crash

After experiencing a total server failure, we are back online. We apologize for the inconvenience - we are still in...


Read More...

L1: motion features

Layer L1 provides an input to the compositional hierarchy. Motion, obtained in L0 is encoded using a small dictionary.


Read More...
01234

State of the art in the proposed field

The research, proposed in this project, is related to two main research areas. The central theme of the proposal is the analysis of visual motion, in particular the analysis of human motion, with a special emphasis on the recognition of actions and activities. There is a large body of literature available on this subject, spreading over various scientific disciplines, from natural sciences to humanities, and with a remarkable contribution from the computer vision field. Due to space constraints, we limit here ourselves only to the most relevant works. An in­depth overview of this field, with a short list of references to the major review papers, is presented in [Aggarwal2011].

Despite significant efforts, there are many remaining challenges in the current state­of­the­ art computer vision­based motion analysis methods. We plan to address those challenges with the introduction of hierarchical compositional models into computer vision­based activity recognition in a similar way than they have been introduced to visual shape analysis and object class categorisation. The approaches, based on the concept of compositionality have become an important research topic in recent years. Therefore an overview of related work on hierarchical compositional models for recognition is also provided.

By tackling the analysis of visual motion through the methodological means of hierarchical compositionality we expect to achieve a significant progress in the area of action recognition and categorization.

This website uses cookies to manage authentication, navigation, and other functions. By using our website, you agree that we can place these types of cookies on your device.

View e-Privacy Directive Documents

You have declined cookies. This decision can be reversed.

You have allowed cookies to be placed on your computer. This decision can be reversed.