From Few to Many: Illumination Cone Models for Face Recognition under Variable Lighting and Pose
IEEE Transactions on Pattern Analysis and Machine Intelligence
Explorations in engagement for humans and robots
Artificial Intelligence
| Hi-index | 0.00 |
Gaze direction is an important communicative cue. In order to use this cue for human-robot interaction, software needs to be developed that enables the estimation of head pose. We began by designing an application that is able to make a good estimate of the head pose, and, contrary to earlier head pose estimation approaches, that works for non-optimal lighting conditions. Initial results show that our approach using multiple networks trained with differing datasets, gives a good estimate of head pose, and it works well in poor lighting conditions and with low-resolution images. We validated our head pose estimation method using a custom built database of images of human heads. The actual head poses were measured using a trakStar (Ascension Technologies) six-degrees-of-freedom sensor. The head pose estimation algorithm allows us to assess a person's focus of attention, which allows robots to react in a timely fashion to dynamic human communicative cues.