Rungsarityotin, W., Starnet, T. "Finding location using omnidirectional video on a wearable computing platform." In Proceedings of ISWC 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....full duplex, only full duplex audio and video with half duplex. They found that help of an expert made the results quicker and more accurate and that there was no media sensitivity in the task. In addition there are wearable computer video applications used for indoor navigation and positioning [3, 4], but which are not relevant to the current application areas. Although there are studies about distance maintenance [5] and case studies from the oil industry [6] they are only of minor relevance because they only partially concern remote inspection. 2. The remote inspection project The main ....

Rungsarityotin, W., Starnet, T. "Finding location using omnidirectional video on a wearable computing platform." In Proceedings of ISWC 2000.

....the user s hands and reaching space, whereas in [9, 13] cameras are strapped to the wearer s hands themselves. In [10] a hat mounted camera looks forward, an orientation also used when the camera is attached to a head mounted display [2] In contrast, 11, 6] uses a camera worn on the chest, in [5] an omnidirectional camera is used above the head, and a wide angle lens camera mounted at the back in [12] and in previous work [4] we placed a miniature active vision system on the shoulder. In a previous paper [4] we identified three frames of reference for measurements that a wearable sensor ....

W. Rungsarityotin and T. Starner, "Finding location using omnidirectional video on a wearable computing platform," IEEE International Symposium on Wearable Computing, Atlanta GA, USA, October, 2000.

.... to a catalogue of previously gathered features for the given environment, making it possible to identify discrete events, such as entering a room [29, 2] Other recent work attempts to identify continuous paths through an environment based on omnidirectional imagery captured from the user s head [27]. Given the wide range of strengths and weaknesses that different tracking technologies have in different circumstances, one promising approach is to combine a set of complementary technologies to create hybrid trackers that are more robust or accurate than any of the individual technologies on ....

W. Rungsarityotin and T. Starner. Finding location using omnidirectional video on a wearable computing platform. In Proc. ISWC '00 (Fourth Int. Symp. on Wearable Computers), pages 61--68, Atlanta, GA, October 16--17 2000.

....systems. Proprietary infrastructure based systems such as GPS [7] RF and ultrasound [8] provide location information with a high infrastructure cost. Also, proprietary non infrastructure based systems such as video based systems utilizing pattern matching techniques exists as described by [3] and [4]. These systems currently need significant amounts of training to achieve accurate results. 2. Approach and Architecture We have developed a five layer architecture for pervasive computer, Handy Andy as shown in Figure 1. The bottom of the figure has a range of mobile and fixed devices. They are ....

Rungsarityotin, W., Starner, T. E., "Finding Location Using Omnidirectional Video on a Wearable Computing Platform." Proceedings of the Fourth International Symposium on Wearable Computers (ISWC'00), Atlanta, GA, 2000.

....context, depending on the amount and complexity of the sensors. The first approach is a well established one where one sensor (or a set of a few sensors) is combined with an algorithm that is usually specific to that particular sensor. Research on wearable cameras with computer vision algorithms [17,20], microphones with sound specific preprocessing [2] or beacon based systems [12] for example give excellent results. The alternative approach is based on a large number of small and simple sensors. This direction was taken in research at MERL [8] Philips [5] and in the TEA project [4,19] but ....

W. Rungsarityotin and T. H. Starner, "Finding location using omnidirectional video on a wearable computing platform". In Proceedings of the Fourth International Symposium on WearableComputers, IEEE Press, pp.61-68, 2000.

....of HMMs for face recognition and detection. However,search at scale was performed and no background or noise models were used. Unfortunately, classic detection is usually under constrained and over optimistic about background content. In examination of 10 standard face databases ( 19, 000 images)[19, 1, 27, 26, 23, 14, 8, 33, 25, 13, 20], we found that background contents had little variation. By comparison, scenes obtained from a body worn camera in everyday life contained highly varied scene backgrounds. Furthermore, current general purpose detectors are very compute intensive due to searching at scale. Though a low false ....

W. Rungsarityotin and T. Starner. Finding location using omnidirectional video on a wearable computing platform. In International Symposium on Wearable Computing, Atlanta, GA, October 2000. IEEE.

No context found.

Rungsarityotin, W. and Starner, T. (2000). Finding location using omnidirectional video on a wearable computing platform. In International Symposium on Wearable Computing, Atlanta, GA. IEEE.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC