Robust Extrinsic Camera Calibration from Trajectories in Human–Populated Environments
Calibración extrínseca robusta de un sistema de cámaras a partir de trayectorias en ambientes humanos
Guillermo Baqueiro Victorín1 and Jean Bernard Hayet2
1 Digipro, Distrito Federal, Mexico. Correo: baqueiro@cimat.mx.
]]> 2 Centro de Investigación en Matemáticas (CIMAT A.C.), Guanajuato, Mexico. Correo: jbhayet@cimat.mx.
Article received on 11/03/2010.
Accepted on 04/02/2011.
Abstract
This paper proposes a novel robust approach to perform inter–camera and ground–camera calibration in the context of visual monitoring of human–populated areas. By supposing that the monitored agents evolve on a single plane and that the cameras intrinsic parameters are known, we use the image trajectories of moving objects as tracked by standard trackers in a RANSAC paradigm to estimate the extrinsic parameters of the different cameras. We illustrate the performance of our algorithm on several challenging experimental setups and compare it to existing approaches.
Keywords: Calibration, computer vision, tracking, video–surveillance and multiple camera systems.
Resumen
]]> Este artículo propone un nuevo método robusto para realizar las calibraciones inter–cámaras y suelo–cámara en el contexto de vídeo–vigilancia sobre escenas pobladas por humanos. Suponemos que los agentes transitan en un simple plano y que los parámetros intrínsecos de las cámaras son conocidos. Usamos las trayectorias de objetos en movimiento en las imágenes, como por ejemplo las generadas por algoritmos de rastreo del estado del arte, para estimar los parámetros extrínsecos de las diferentes cámaras. Ilustramos el desempeño de nuestro algoritmo sobre diferentes configuraciones experimentales desafiantes, y lo comparamos con diferentes métodos existentes.Palabras clave: Calibración de cámaras, visión por computadora, rastreo, vídeo–vigilancia y sistemas de cámaras múltiples.
DESCARGAR ARTÍCULO EN FORMATO PDF
References
1. Caspi, Y., Símakov, D., & Iraní, M. (2006). Feature–Based Sequence–to–Sequence Matching. International Journal of Computer Vision, 68(1), 5364. [ Links ]
2. Chum, O., Matas, J., & Kittler, J. (2003). Locally Optimized RANSAC. 25th DAGM Symposium, Lecture notes in Computer Science, 2781, 236243. [ Links ]
]]>3. Du, W., Hayet, J.B., Verly, J. & Piater, J. (2009). Ground–Target Tracking in Multiple Cameras Using Collaborative Particle Filters and Principal Axis–Based Integration. IPSJ Transactions on Computer Vision and Applications, 1, 58–71. [ Links ]
4. Eleventh IEEE International Workshop on Performance Evaluation of Tracking and Surveillance. (PETS 2009). Retrieved from: http://www.cvg.rdg.ac.uk/PETS2009/ [ Links ]
5. Hartley, R. & Zisserman, A. (2000). Multiple View Geometry in Computer Vision, Cambridge, UK: Cambridge University Press. [ Links ]
6. Kayumbi, G. & Cavallaro, A. (2008). Multi–view trayectory mapping using homography with lens distortion correction. EURASIP Journal on Image and Video Processing, 2008, Article ID 145715. [ Links ]
7. Lee, L., Romano, R., & Stem, G. (2000). Monitoring Activities from Multiple Video Streams: Establishing a Common coordinate Frame. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(8), 758–767. [ Links ]
8. Mikolajczyk, K. & Schmid, C. (2005). A performance evaluation of local descriptors. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(10), 1615–1630. [ Links ]
9. Nunziati, W., Sclaroff, S., & Del Bimbo, A. (2010). Matching Trayectories between Video Sequences by Exploiting a Sparse Proyective Invariant Representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(3), 517–529. [ Links ]
10. Stauffer, C. & Thieu, K. (2003). Automated multi–camera planar tracking correspondence modelling. 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Madison, WI, USA, 259–266. [ Links ]
11. Triggs, B. (1998). Autocalibration from Planar Scenes. 5th European Conference in Computer Vision (ECCV'98), Freiburg, Germany, 1, 89–105. [ Links ]
]]>