Online Recognition of Actions Involving Objects
This work addresses the problem of online action-object recognition for robotics or surveillance systems, but it appears incremental as it combines existing techniques like SOMs and proximity measures without introducing a fundamentally new approach.
The authors tackled the problem of real-time action recognition involving objects by merging two parallel information streams: one for recognizing actions from spatial trajectories using hierarchical self-organizing maps and a supervised neural network, and another for identifying the target object via proximity measures, achieving excellent performance in tests.
We present an online system for real time recognition of actions involving objects working in online mode. The system merges two streams of information processing running in parallel. One is carried out by a hierarchical self-organizing map (SOM) system that recognizes the performed actions by analysing the spatial trajectories of the agent's movements. It consists of two layers of SOMs and a custom made supervised neural network. The activation sequences in the first layer SOM represent the sequences of significant postures of the agent during the performance of actions. These activation sequences are subsequently recoded and clustered in the second layer SOM, and then labeled by the activity in the third layer custom made supervised neural network. The second information processing stream is carried out by a second system that determines which object among several in the agent's vicinity the action is applied to. This is achieved by applying a proximity measure. The presented method combines the two information processing streams to determine what action the agent performed and on what object. The action recognition system has been tested with excellent performance.