Action recognition based on hierarchical self-organizing maps

We propose a hierarchical neural architecture able to recognise observed human actions. Each layer in the architecture represents increasingly complex human activity features. The first layer consists of a SOM which performs dimensionality reduction and clustering of the feature space. It represents the dynamics of the stream of posture frames in action sequences as activity trajectories over time. The second layer in the hierarchy consists of another SOM which clusters the activity trajectories of the first-layer SOM and thus it learns to represent action prototypes independent of how long the activity trajectories last. The third layer of the hierarchy consists of a neural network that learns to label action prototypes of the second-layer SOM and is independent - to certain extent - of the camera's angle and relative distance to the actor. The experiments were carried out with encouraging results with action movies taken from the INRIA 4D repository. The architecture correctly recognised 100% of the actions it was trained on, while it exhibited 53% recognition rate when presented with similar actions interpreted and performed by a different actor.