Object Activity Scene Description, Construction and Recognition
This work addresses activity scene recognition for social robots, but it is incremental as it builds on existing methods like motion attention and CNNs.
The paper tackles the challenge of recognizing a group of actions in an activity scene by partitioning scenes into primitive actions using motion attention, describing them with joint trajectory vectors, and applying CNN for recognition, achieving efficiency on a human activity dataset.
Action recognition is a critical task for social robots to meaningfully engage with their environment. 3D human skeleton-based action recognition is an attractive research area in recent years. Although, the existing approaches are good at action recognition, it is a great challenge to recognize a group of actions in an activity scene. To tackle this problem, at first, we partition the scene into several primitive actions (PAs) based upon motion attention mechanism. Then, the primitive actions are described by the trajectory vectors of corresponding joints. After that, motivated by text classification based on word embedding, we employ convolution neural network (CNN) to recognize activity scenes by considering motion of joints as "word" of activity. The experimental results on the scenes of human activity dataset show the efficiency of the proposed approach.