Cloud Robotics Based on Facial Attribute Analysis for Human-Robot Interaction
- Yuji Yamauchi, Yu Kato, Takayoshi Yamashita, and Hironobu Fujiyoshi
- IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN)
In this paper, we propose a facial attribute image analysis engine for cloud robotics. Human support robots need to utilize a variety of recognition technologies because of the many tasks they must perform. These can be divided into two categories: recognizing persons and recognizing objects. In this paper, we describe a system that focuses on the former. To achieve natural communication between persons and robots, it is necessary to observe the person’s intention and physical states. It is also necessary to observe these for early disease detection and safety confirmation. For this we need to implement the cloud server with a recognition engine that can recognize persons from their facial images and attributes (facial points, race, age, gender, facial expressions, etc.).