Suivi par ré-identification dans un réseau de caméras à champs disjoints

Suivi par ré-identification dans un réseau de caméras à champs disjoints

Boris Meden Frédéric Lerasle  Patrick Sayd 

CEA, LIST Laboratoire Vision et Ingénierie des Contenus BP 94, F-91191 Gif-sur-Yvette

CNRS ; LAAS Université de Toulouse ; UPS, LAAS F-31077 Toulouse cedex 4

Corresponding Author Email: 
{boris.meden,patrick.sayd}@cea.fr
Page: 
283-305
|
DOI: 
https://doi.org/10.3166/TS.29.283-305
Received: 
N/A
|
Accepted: 
N/A
|
Published: 
31 August 2012
| Citation

OPEN ACCESS

Abstract: 

This article tackles the problem of automatic multi-pedestrian tracking in non overlapping fields of view camera networks, using monocular, uncalibrated cameras. Tracking is locally addressed by a Tracking-by-Detection and reidentification algorithm. We propose here to introduce the concept of global identity into a multi-target tracking algorithm, qualifying people at the network level, to allow us to rebound observation discontinuities. We embed that identity into the tracking loop thanks to the mixed-state particle filter framework, thus including it in the search space. Doing so, each tracker maintains a mutli-modality on the identity in the network of its target. We increase the decision strength introducing a high level decision scheme which integrates all the trackers hypothesis over all the cameras of the network with previous reidentification results and the topology of the network. The tracking and reidentification module is first tested with a single camera. We then evaluate the whole framework on a 3non-overlapping fields of views network with 7 identities. The only a priori knowledge assumed is a topological map of the network.

RÉSUMÉ

Cet article pose le problème du suivi automatique de piétons à travers les réseaux de caméras à champs de vue disjoints. Le suivi dans l’image est traité de manière locale par un algorithme de suivi par détections et ré-identification. Avec du filtrage particulaire à état continu et discret, nous introduisons la notion d’identité globale dans un algorithme de suivi multipiste pour caractériser les personnes au niveau du réseau et pallier les discontinuités d’observations. Ceci permet à chaque traqueur d’inclure l’identité de la cible qu’il est en train de suivre dans l’espace de recherche. Ce faisant, chaque traqueur maintient à jour une distribution de probabilité discrète sur l’identité de la piste qu’il est en train de suivre. La décision de ré-identification est renforcée par un schéma décisionnel haut niveau intégrant les hypothèses de chaque traqueur confrontées à la topologie du réseau. La composante suivi multipersonne et ré-identification est d’abord testée en contexte monocaméra. Nous évaluons ensuite notre approche complète sur un réseau de 3 caméras à champs de vue disjoints et un ensemble de 7 personnes. La seule connaissance a priori requise est la carte topologique du réseau.

Keywords: 

re-identification, pedestrian tracking, camera network, nonoverlapping fields of view, particle filtering

MOTS-CLÉS

ré-identification, suivi de personnes, réseau de cameras, champs de vue disjoints, filtrage particulaire

Extended Abstract
1. Introduction
2. État De L’art
3. Suivi Par Ré-Identification Au Sein D’une Caméra
4. Supervision Topologico-Temporelle Des Ré-Identifications
5. Implémentation Et Évaluations Associées
6. Conclusion Et Perspectives
  References

Bar-Shalom Y., Fortmann T., Scheffe M. (1980). Joint probabilistic data association for multiple targets in clutter. In Proc. conf. on information sciences and systems, p. 404–409.

Benfold B., Reid I. (2011). Stable multi-target tracking in real-time surveillance video. In Proceedings of the international conference on computer vision and pattern recognition.

Bernardin K., Stiefelhagen R. (2008). Evaluating multiple object tracking performance: the clear mot metrics. Journal on Image and Video Processing.

Breitenstein M., Reichlin F., Leibe B., Koller-Meier E., Van Gool L. (2010). Online multiperson tracking-by-detection from a single, uncalibrated camera. IEEE Transactions on Pattern Analysis and Machine Intelligence.

Burgeois F., Lasalle J.-C. (1971). An extension of the munkres algorithm for the assignment problem to rectangular matrices. Communications of the ACM.

Chen K., Lai C., Hung Y., Chen C. (2008). An adaptive learning method for target tracking across multiple cameras. In Proceedings of the international conference on computer vision and pattern recognition.

Cheng D. S., Cristani M., Stoppa M., Bazzani L., Murino V. (2011). Custom pictorial structures for re-identification. In Proceedings of the british machine vision conference.

Cox I., Hingorani S. (1996). An efficient implementation of reid’s multiple hypothesis tracking algorithm and its evaluation for the purpose of visual tracking. Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 18, no 2, p. 138–150.

Dalal N., Triggs B. (2005). Histograms of oriented gradients for human detection. In Proceedings of the international conference on computer vision and pattern recognition.

Danchick R., Newnam G. (2006). Reformulating reid’s mht method with generalised murty kbest ranked linear assignment algorithm. In Radar, sonar and navigation, iee proceedings-, vol. 153, p. 13–22.

Ess A., Leibe B., Van Gool L. (2007). Depth and appearance for mobile scene analysis. In Proceedings of the international conference on computer vision, p. 1–8.

Farenzena M., Bazzani L., Perina A., Murino V., Cristani M. (2010). Person re-identification by symmetry-driven accumulation of local features. In Proceedings of the international conference on computer vision and pattern recognition.

Forssén P. (2007). Maximally stable colour regions for recognition and matching. In Proceedings of the international conference on computer vision and pattern recognition, p. 1–8.

Gheissari N., Sebastian T., Hartley R. (2006). Person reidentification using spatiotemporal appearance. In Proceedings of the international conference on computer vision and pattern recognition.

Gray D., Brennan S., Tao H. (2007). Evaluating appearance models for recognition, reacquisition, and tracking. In Proc. ieee international workshop on performance evaluation for tracking and surveillance (pets).

Gray D., Tao H. (2008). Viewpoint invariant pedestrian recognition with an ensemble of localized features. In Proceedings of the european conference on computer vision.

Huang T., Russell S. (1997). Object identification in a bayesian context. In Proceedings of the international joint conference on artificial intelligence.

Isard M., Blake A. (1998a). Condensation-conditional density propagation for visual tracking. International journal of computer vision, vol. 29, no 1, p. 5–28.

Isard M., Blake A. (1998b). A mixed-state CONDENSATION tracker with automatic modelswitching. In Proceedings of the international conference on computer vision.

Isard M., Blake A. (2001). BraMBLe: a Bayesian multiple blob tracker. In Proceedings of the international conference on computer vision.

Javed O., Shafique K., Shah M. (2005). Appearance modeling for tracking in multiple nonoverlapping cameras. In Proceedings of the international conference on computer vision and pattern recognition.

Kaucic R., Perera A., Brooksby G., Kaufhold J., Hoogs A. (2005). A unified framework for tracking through occlusions and accross sensor gaps. In Proceedings of the international conference on computer vision and pattern recognition.

Kettnaker V., Zabih R. (1999). Bayesian multi-camera surveillance. In Proceedings of the international conference on computer vision and pattern recognition.

Kuo C., Huang C., Nevatia R. (2010). Inter-camera association of multi-target tracks by on-line learned appearance affinity models. In Proceedings of the european conference on computer vision.

Kuo C., Nevatia R. (2011). How does person identity recognition help multi-person tracking? In Proceedings of the international conference on computer vision and pattern recognition.

Lev-Tov A., Moses Y. (2010). Path recovery of a disappearing target in a large network of cameras. In Proceedings of the international conference on distributed smart cameras.

Makris D., Ellis T., Black J. (2004). Bridging the gaps between cameras. In Proceedings of the international conference on computer vision and pattern recognition.

Matei B., Sawhney H., Samarasekera S. (2011). Vehicle tracking across nonoverlapping cameras using joint kinematic and appearance features. In Proceedings of the international conference on computer vision and pattern recognition.

Meden B., Sayd P., Lerasle F. (2011). Mixed-State Particle Filtering for Simultaneous Tracking and Re-Identification in Non-Overlapping Camera Networks. In Proceedings of the scandinavian conference on image analysis (scia). Ystad, Suède.

Meden B., Sayd P., Lerasle F. (2012). Suivi par ré-identification dans un réseau de caméras à champs disjoints. In Actes du congrès francophone sur la reconnaissance des formes et l’intelligence artificielle (rfia). Lyon.

Oh S., Russell S., Sastry S. (2004). Markov chain monte carlo data association for general multiple-target tracking problems. In Cdc.

Okuma K., Taleghani A., De Freitas N., Little J., Lowe D. (2004). A boosted particle filter: multitarget detection and tracking. In Proceedings of the european conference on computer vision.

Oreifej O., Mehran R., Shah M. (2010). Human identity recognition in aerial images. In Proceedings of the international conference on computer vision and pattern recognition.

Pasula H., Russell S., Ostland M., Ritov Y. (1999). Tracking many objects with many sensors. In Proceedings of the international joint conference on artificial intelligence.

Prisacariu V., Reid I. (2009). fasthog - a real-time gpu implementation of hog. Rapport technique no 2310/09. Department of Engineering Science, Oxford University.

Prosser B., Zheng W., Gong S., Xiang T., Mary Q. (2010). Person Re-Identification by Support Vector Ranking. In Proceedings of the british machine vision conference.

Qu W., Schonfeld D., Mohamed M. (2007). Distributed bayesian multiple-target tracking in crowded environments using multiple collaborative cameras. Int. Journal EURASIP.

Reid D. (1979). An algorithm for tracking multiple targets. Automatic Control, IEEE Transactions on, vol. 24, n° 6, p. 843-854.

Wojek C., Roth S., Schindler K., Schiele B. (2010). Monocular 3D scene modeling and inferences: understanding multi-object traffic scenes. Proceedings of the european conference on computer vision.

Xing J., Ai H., Lao S. (2009). Multi-object tracking through occlusions by local tracklets filtering and global tracklets association with detection responses. Proceedings of the international conference on computer vision and pattern recognition, p. 1200-1207.

Zajdel W., Kröse B. (2005). A sequential bayesian algorithm for surveillance with nonoverlapping cameras. International Journal of Pattern Recognition and Artificial Intelligence.

Zheng W., Gong S., Xiang T. (2009). Associating groups of people. Proceedings of the british machine vision conference, vol. 7.

Zheng W., Gong S., Xiang T. (2011). Person re-identification by probabilistic relative distance comparison. In Proceedings of the international conference on computer vision and pattern recognition.