Object Recognition for the Internet of Things
We present a system which allows to request information on physical objects by taking a picture of them. This way, using a mobile phone with integrated camera, users can interact with objects or ”things” in a very simple manner. A further advantage is tha
- PDF / 7,250,704 Bytes
- 17 Pages / 430 x 660 pts Page_size
- 22 Downloads / 220 Views
Abstract. We present a system which allows to request information on physical objects by taking a picture of them. This way, using a mobile phone with integrated camera, users can interact with objects or ”things” in a very simple manner. A further advantage is that the objects themselves don’t have to be tagged with any kind of markers. At the core of our system lies an object recognition method, which identifies an object from a query image through multiple recognition stages, including local visual features, global geometry, and optionally also metadata such as GPS location. We present two applications for our system, namely a slide tagging application for presentation screens in smart meeting rooms and a cityguide on a mobile phone. Both systems are fully functional, including an application on the mobile phone, which allows simplest point-andshoot interaction with objects. Experiments evaluate the performance of our approach in both application scenarios and show good recognition results under challenging conditions.
1
Introduction
Extending the Internet to physical objects - the Internet of Things - promises humans to live in a smart, highly networked world, which allows for a wide range of interactions with this environment. One of the most convenient interactions is the request of information about physical objects. For this purpose several methods are currently being discussed. Most of them rely on some kind of unique marker integrated in or attached to the object. Some of these markers can be analyzed using different kinds of wireless near field communication (for instance RFID tags [24] or Bluetooth beacons [11]), others are visual markers and can be analyzed using cameras, for instance standard 1D-barcodes [2] or their modern counterparts, the 2D codes [21]. A second development concerns the input devices for interaction with physical objects. In recent years mobile phones have become sophisticated multimedia computers that can be used as flexible interaction devices with the user’s environment. Besides the obvious telephone capabilities, current devices offer integrated cameras and a wide range of additional communication channels such as Bluetooth, WLAN or access to the Internet. People are used to the device they own and usually carry it with them all day. Furthermore, with the phone-number, C. Floerkemeier et al. (Eds.): IOT 2008, LNCS 4952, pp. 230–246, 2008. c Springer-Verlag Berlin Heidelberg 2008
Object Recognition for the Internet of Things
231
a device is already tied to a specific person. Thus it is only natural to use the mobile phone as a personal input device for the Internet of things. Indeed, some of the technologies mentioned above have already been integrated in mobile phones, for instance barcode readers or RFID readers. The ultimate system, however, would not rely on markers to recognize the object, but rather identify it by its looks, i.e. using visual object recognition from a mobile phone’s camera image. Since the large majority of mobile phones contain an integrated camera, a
Data Loading...
