Abstract—Human action and activity recognition from videos has attracted an increasing number of researchers in recent years. However, most of the works aim at multimedia retrieval and surveillance applications, but rarely at humanoid household robots, even though the robotic perception of human activities would allow a more natural human-robot interaction (HRI). To encourage future studies in this domain, we present in this work a novel data set specifically designed for the application in HRI scenarios. This Robo-kitchen data set consists of 14 typical kitchen activities recorded in two different stereo-camera setups, and each performed by 17 subjects. To establish a baseline for future work, we extend a state-of-the-art action recognition method to be applicable on the activity classification problem and evaluate it on the Robo-kitchen data set showing promising results. I.

Abstract. In robotics research is an increasing need for knowledge about human motions. However humans tend to perceive motion in terms of discrete motion primitives. Most systems use data-driven motion segmentation to retrieve motion primitives. Besides that the actual intention and context of the motion is not taken into account. In our work we propose a procedure for segmenting motions according to their functional goals, which allows a structuring and modeling of functional motion primitives 3. The manual procedure is the first step towards an automatic functional motion representation. This procedure is useful for applications such as imitation learning and human motion recognition. We applied the proposed procedure on several motion sequences and built a motion recognition system based on manually segmented motion capture data. We got a motion primitive error rate of 0.9 % for the marker-based recognition. Consequently the proposed procedure yields motion primitives that are suitable for human motion recognition. 1