A novel interactive exhibition method based on handheld devices

We propose a novel interactive method for museum exhibitions based on handheld devices such as ordinary camera-equipped mobile phones. It enables museum visitors to automatically get annotations about exhibits interest in a non-distractive way by capturing videos of them. Our method recognizes different comment region of exhibits by applying SIFT algorithm on handheld devices. While SIFT is known to be a strong algorithm for local feature recognition, it is computationally expensive and unsuitable for low-end devices such as mobile phones. We introduce some modified approaches to make it suitable for handheld devices. The usefulness of the proposed method and user satisfaction are evaluated through the experimental results.