S. Utcke. Grouping based on projective geometry constraints and uncertainty. Internal Report 1/97, Technische Informatik I, TU-HH, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....real time recognition of landmarks As demonstration, the real time recognition has been carried out with the above described system of blind glasses for the selected landmarks, including telephone booth, mailbox (see Fig. 6(a) and (b) and pedestrian crossing (separately described in [10] by Sven Utcke) The images are taken by the blind glasses, recognized by the parallel computer and reported by a microphone to people on the site. The experiments prove that the Figure 5: a) stationary (b) mobile component of blind glasses recognition is stable and robust with the above ....

S. Utcke, Grouping based on projective geometry constraints and uncertainty, ICCV'98, Bambay, India, January 1998, pp739-746, Narosa Publishing House.

....a range of indoor and outdoor images. We also show that estimating the grid structure makes it significantly easier to detect target objects which are not aligned with the grid. 1 Introduction Recently there has been growing interest in building computer vision navigational systems for the blind [9], 10] These systems can be used, for example, for navigation and for the detection and reading of informational signs. The goal of this paper is to determine the orientation of the viewer in the scene (indoor or outdoor) from a single image. A useful spin off is the ability to detect target ....

....modulo 90 ffi . 2 Previous Work and Three Dimensional Geometry There has been an enormous amount of work in projective geometry [3, 6] Techniques from projective geometry have been applied to finding the vanishing points [1] 5] For a recent application to vision systems for the blind see [9] for the detection of pedestrian crossings using projection geometry. This work, however, has typically proceeded through the stages of edge detection, Hough transforms, and finally the calculation of the geometry. Alternatively, a sequence of images over time can be used to estimate the geometry, ....

S. Utcke. "Grouping based on Projective Geometry Constraints and Uncertainty". In Proceedings of the International Conference on Computer Vision. ICCV'98. Bombay, India. pp 739-746. 1998.

....group consisting of relatively small number of lines will be left undetected with this approach. 3.2 The Detection Algorithm A di erent approach is proposed here where potential groups of candidate lines are generated and then tested for coincidence. This approach was employed by Utcke [25] for grouping and recognising zebra crossing in cluttered images. Line clustering was also used in [15] to classify groups which share common vanishing points followed by vanishing point estimation. Based on the projective property of structures with parallel lines, our algorithm [17] picks out ....

S. Utcke. Grouping based on projective geometry constraints and uncertainty. In Proceedings of the Sixth International Conference on Computer Vision, pages 739-746, Bombay, India, January 1998.

....20 down to 5. In fact, any convergent group consisting of relatively small number of lines will be left undetected with this approach. 2.2 The Detection Algorithm A di erent approach is proposed here where potential groups of candidate lines are generated and then tested for coincidence. Utcke [21] used a similar approach for grouping and recognizing zebra crossing taking into account image feature uncertainties, crossratio and vanishing line constraints. This approach was employed to detect stair cases in [14, 19] Hough Transform line tting is applied to the image rst. Based on the ....

S. Utcke. Grouping based on projective geometry constraints and uncertainty. In Proceedings of the Sixth International Conference on Computer Vision, pages 739-746, Bombay, India, January 1998.

....of multiple groupings and clutter features. The response of a grouping algorithm indicates the likely presence or absence of an object in a scene [8,25] In our case, the response of any of the groupers that we presented is evidence for the presence of things like buildings, zebra crossings [24] and brick walls. Such groupings define a feature that may be used as the basis for further tasks such as image matching, and this is the subject of current work. Acknowledgements The algorithms in this paper were implemented using the IUE TargetJr software packages. This work was supported by ....

S. Utcke. Grouping based on projective geometry constraints and uncertainty. In Proc. ICCV, Jan 1998.

....to calculate a meaningful cross ratio even if the four lines are not quite coincident due to image noise. The work on uncertainty described here owes much to work by Kanatani [5 7] A more detailed comparison with the literature, which has been omitted due to space limitations, can be found in [11]. 2 Uncertainty This section tries to answer the questions why modelling uncertainty should be beneficial and how. Section 2.1 discusses how exactly knowledge about a result s uncertainty can help to reach a well founded decision, while Section 2.2 answers the question how to obtain knowledge ....

S. Utcke. Grouping based on projective geometry constraints and uncertainty. Internal Report 1/97, Technische Informatik I, TU-HH, 1997.

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