Automatic occlusion removal from facades for 3D urban reconstruction
Authors: 
Chris Engels, David Tingdahl, Mathias Vercruysse, Tinne Tuytelaars, Hichem Sahli, Luc Van GoolAuthors Info & Claims
Chris Engels, David Tingdahl, Mathias Vercruysse, Tinne Tuytelaars, Hichem Sahli, Luc Van GoolAuthors Info & ClaimsPages 681 - 692
Abstract
Object removal and inpainting approaches typically require a user to manually create a mask around occluding objects. While creating masks for a small number of images is possible, it rapidly becomes untenable for longer image sequences. Instead, we accomplish this step automatically using an object detection framework to explicitly recognize and remove several classes of occlusions. We propose using this technique to improve 3D urban reconstruction from street level imagery, in which building facades are frequently occluded by vegetation or vehicles. By assuming facades in the background are planar, 3D scene estimation provides important context to the inpainting process by restricting input sample patches to regions that are coplanar to the occlusion, leading to more realistic final textures. Moreover, because non-static and reflective occlusion classes tend to be difficult to reconstruct, explicitly recognizing and removing them improves the resulting 3D scene.
References
[1]
Barnes, C., Shechtman, E., Finkelstein, A., Goldman, D.B.: PatchMatch: A randomized correspondence algorithm for structural image editing. ACM Transactions on Graphics (Proc. SIGGRAPH) 28(3) (August 2009)
[2]
Benitez, S., Denis, E., Baillard, C.: Automatic production of occlusion-free rectified facade textures using vehicle-based imagery. In: Photogrammetric Computer Vision and Image Analysis, p. A:275 (2010)
[3]
Boykov, Y., Kolmogorov, V.: An experimental comparison of min-cut/max- flow algorithms for energy minimization in vision. IEEE Transactions on Pattern Analysis and Machine Intelligence 26(9), 1124-1137 (2004)
[4]
Criminisi, A., Perez, P., Toyama, K.: Region filling and object removal by exemplarbased image inpainting. IEEE Transactions on Image Processing 13, 1200-1212 (2004)
[5]
Dick, A.R., Torr, P.H.S., Cipolla, R.: Modelling and interpretation of architecture from several images. Int. J. Comput. Vision 60, 111-134 (2004)
[6]
Felzenszwalb, P., Girshick, R., McAllester, D.: Cascade object detection with deformable part models. In: Computer Vision and Pattern Recognition (2010)
[7]
Hartley, R.I., Zisserman, A.: Multiple View Geometry in Computer Vision, 2nd edn. Cambridge University Press, Cambridge (2004) ISBN: 0521540518
[8]
Hedau, V., Hoiem, D., Forsyth, D.: Recovering the spatial layout of cluttered rooms. In: International Conference on Computer Vision (2009)
[9]
Hoiem, D., Efros, A.A., Hebert, M.: Putting objects in perspective. International Journal of Computer Vision 80(1), 3-15 (2008)
[10]
Kazhdan, M., Bolitho, M., Hoppe, H.: Poisson surface reconstruction. In: Proceedings of the Fourth Eurographics Symposium on Geometry Processing, SGP 2006, pp. 61-70. Eurographics Association, Aire-la-Ville (2006)
[11]
Konushin, V., Vezhnevets, V.: Abstract automatic building texture completion. Graphicon (2007)
[12]
Leibe, B., Leonardis, A., Schiele, B.: Robust object detection with interleaved categorization and segmentation. International Journal of Computer Vision 77(1- 3), 259-289 (2008)
[13]
Rasmussen, C., Korah, T., Ulrich, W.: Randomized view planning and occlusion removal for mosaicing building facades. In: IEEE International Conference on Intelligent Robots and Systems (2005), http://nameless.cis.udel.edu/pubs/2005/ RKU05
[14]
Rother, C., Kolmogorov, V., Blake, A.: "grabcut": interactive foreground extraction using iterated graph cuts. ACM Trans. Graph. 23, 309-314 (2004)
[15]
Saxena, A., Chung, S.H., Ng, A.Y.: 3-d depth reconstruction from a single still image. International Journal of Computer Vision, IJCV 76 (2007)
[16]
Thomas, A., Ferrari, V., Leibe, B., Tuytelaars, T., Van Gool, L.: Shape-fromrecognition: Recognition enables meta-data transfer. Computer Vision and Image Understanding 113(12), 1222-1234 (2009)
[17]
Thomas, A., Ferrari, V., Leibe, B., Tuytelaars, T., Schiele, B., Van Gool, L.: Towards multi-view object class detection. In: Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006, vol. 2, pp. 1589-1596. IEEE Computer Society, Washington, DC, USA (2006)
[18]
Vergauwen, M., Van Gool, L.: Web-based 3d reconstruction service. Mach. Vision Appl. 17(6), 411-426 (2006)
[19]
Wang, L., Jin, H., Yang, R., Gong, M.: Stereoscopic inpainting: Joint color and depth completion from stereo images. In: Conference on Computer Vision and Pattern Recognition (2008)
[20]
Werner, T., Zisserman, A.: Model selection for automated reconstruction from multiple views. In: British Machine Vision Conference, pp. 53-62 (2002)
[21]
Werner, T., Zisserman, A.: New techniques for automated architectural reconstruction from photographs. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2351, pp. 541-555. Springer, Heidelberg (2002)
[22]
Xiao, J., Fang, T., Zhao, P., Lhuillier, M., Quan, L.: Image-based street-side city modeling. ACM Trans. Graph. 28, 114:1-114:12 (2009)
- Automatic occlusion removal from facades for 3D urban reconstruction
Recommendations
Occlusion Detectable Stereo -- Occlusion Patterns in Camera Matrix
CVPR '96: Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)In stereo algorithms with more than two cameras, the improvement of accuracy is often reported since they are robust against noise. However, another important aspect of the polynocular stereo, that is the ability of occlusion detection, has been paid ...
3D Reconstruction of an Urban Scene from Synthetic Fish-Eye Images
SSIAI '00: Proceedings of the 4th IEEE Southwest Symposium on Image Analysis and Interpretation3D scene reconstruction from 2D stereo pairs or sequences has been an interesting issue in computer vision for years. So far, most research has used images acquired by linear low-distortion lenses that can be modeled by pinhole mapping. In recent years, ...
Monocular 3D scene understanding with explicit occlusion reasoning
CVPR '11: Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern RecognitionScene understanding from a monocular, moving camera is a challenging problem with a number of applications including robotics and automotive safety. While recent systems have shown that this is best accomplished with a 3D scene model, handling of ...
Comments
Information & Contributors
Information
Published In
August 2011
759 pages
ISBN:9783642236860
Sponsors
- Alcatel-Lucent
- Ghent University - Faculty of Engineering and Architecture: Ghent University - Faculty of Engineering and Architecture
- Flemish Fund for Scientific Research - FWO-Vlaanderen: Flemish Fund for Scientific Research - FWO-Vlaanderen
- Philips: Philips
- IBBT: Interdisciplinary Institute for Broadband Technology
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 22 August 2011
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 25 Feb 2025