jected to an image (red 
  
sreen crosses) 
   
  
een point) of image I and 
| point) 
  
  
  
  
  
  
  
  
  
  
    
  
  
   
    
    
   
RRRERURRS 
ge 
AERA 
FRIAR 
URRRRRREM 
   
Figure 7b: Images used to reconstruct the scene 
  
1 
® 
ET E. 
i 
= 
P 
Hin 
Em 
» 
Ep em a d 
TH 
gh s mn an din 
ju 
* 
aum uns 
uu a s on 8 
FIT LM 
Emma" 
  
  
Figure 7c: The estimated viewpoint and its actual place 
4. CONCLUSIONS AND FUTURE WORK 
In this paper we present an approach to solve the Location 
Determination Problem in urban environment using image 
sequences. The outdoor scene depicted in the given image is 
reconstructed first. Points of images used in the reconstruction 
are transferred to the given image in order to obtain the control 
points. Then pose of the given image is estimated using PnP 
solver. The computation of the presented method mainly lies in 
matching and reconstruction process. In the future work, GPU 
computation should be considered to speed up the matching 
process. We may also try other feature operators such as 
Speeded Up Robust Features (SURF) and ORiented Brief (ORB) 
to evaluate their performance in our research environment. We 
will test patch-based algorithms for methods that find both 
dense and global matches have often had high time cost in the 
matching stage. In Parallel Tracking and Mapping (PTAM) 
there are no descriptors as in SIFT but *warped" patches which 
makes it fast and detectable at bigger angles, which makes it 
possible to register images with relative large angles between 
viewpoints. Another function that is worthy to add to our 
method is to geo-locate the reconstructions and the given image 
I as well if the image sequences come with geo-tags/GPS 
information. However, geographical information obtained from 
images is frequently incorrect, noisy and even missing, which 
means we must introduce robust estimation method to further 
improve the accuracy and automation of the presented method. 
Acknowledgement 
This research is supported by the National Natural Science 
Foundation of China (No. 40871211). 
References 
Agarwal, S., Furukawa, Y., Snavely, N., Simon, L, Curless, B., 
Seitz S.M. and Szeliski, R., 2011. Building rome in a day. 
Communications of the ACM, 54(10), pp. 105-112. 
Fischler, M.A. and Bolles, R.C., 1981. Random sample 
consensus: a paradigm for model fitting with applications to 
image analysis and automated cartography. Communications of 
the ACM, 24(6), pp. 381-395. 
Fitzgibbon, A.W. and Zisserman, A., 1998. Automatic camera 
recovery for closed or open image sequences. In: Proceedings 
of the 5th European Conference on Computer Vision, Freiburg, 
Germany, pp. 311-326. 
Hartley, R.I. and Zisserman, A., 2004. Multiple View Geometry 
in Computer Vision. Cambridge University Press, Cambridge, 
pp. 379-383. 
Lepetit, V., Moreno-Noguer, F. and Fua, P., 2009. EPnP: an 
accurate O(n) solution to the PnP problem. International 
Journal of Computer Vision, 81(2), pp. 155-166. 
Lowe D.G., 1999. Object recognition from local scale-invariant 
features. In: Proceedings of the 7th International Conference on 
Computer Vision, Liége, Belgium, pp. 1150-1157. 
Lowe D.G., 2004. Distinctive image features from scale- 
invariant keypoints. International Journal of Computer Vision, 
60(2), pp. 91-110. 
Moreno-Noguer, F., Lepetit, V. and Fua, P., 2007. Accurate 
non-iterative O(n) solution to the PnP problem. In: Proceedings 
of 11th IEEE International Conference on Computer Vision, 
Rio de Janeiro, Brazil, pp. 1-8. 
Snavely, N., Seitz, S.M. and Szeliski, R., 2006. Photo tourism: 
exploring image collections in 3D. ACM Transactions on 
Graphics, 25(3), pp. 835-846. 
Snavely, N., Seitz S.M. and Szeliski, R., 2008. Modeling the 
world from internet photo collections. International Journal of 
Computer Vision, 80(2), pp. 189-210. 
Zhang W. and Kosecka, J., 2006. Image Based Localization in 
Urban Environments. In: Proceedings of the Third 
International Symposium on 3D Data Processing, Visualization, 
and Transmission, North Carolina, Chapel Hill, USA.