Abstract
Pedestrians do not walk randomly. While they move toward their desired destination, they avoid static obstacles and other pedestrians. At the same time they try not to slow down too much as well as not to speed up excessively. Studies coming from the field of social psychology show that pedestrians exhibit common behavioral patterns. For example the distance at which one individual keeps himself from others is not uniformly random, but depends on the acquaintance level of the individuals, the culture and other factors. Our goal here is to use this knowledge to build a model that probabilistically represents the future state of a pedestrian trajectory. To this end, we focus on a stochastic motion model that caters for the possible behaviors in an entire scene in a multi-hypothesis approach, using a principled modeling of uncertainties.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
As in physics, we use the term velocity for a two-dimensional motion vector, as opposed to the scalar value speed.
- 2.
To reduce notational complexity, we will omit the dependency on u, r, O in the rest of the chapter. These values do not change in the time window of the prediction and are assumed to be known.
- 3.
In the original model, the desired velocity is linearly filtered for smoothness (see (14) in [14]). Here, we use an equivalent energy potential that includes already the same smoothing, by introducing a simple coordinate transformation:
$$E\bigl(\mathbf{v}^{t};\mathbf{S}^{t-1}\bigr)=E_{\mathrm{LTA}}\biggl(\frac{\mathbf {v}^{t}-\alpha*\mathbf{v}^{t-1}}{1-\alpha};S^{t-1}\biggr),$$where E LTA is the formulation of the energy given in [14]. Note that this is an entirely equivalent formulation, but it has the advantage of being more compact.
- 4.
We assumed that K and N are time independent and the same for all the subjects. This need not to be the case. However, we drop the dependencies for the sake of readability. The generalization is straightforward.
References
Ali, S., Shah, M.: Floor fields for tracking in high density crowd scenes. In: ECCV (2008)
Antonini, G., Martinez, S.V., Bierlaire, M., Thiran, J.P.: Behavioral priors for detection and tracking of pedestrians in video sequences. Int. J. Comput. Vis. 69, 159–180 (2006)
Arulampalam, M.S., Maskell, S., Gordon, N., Clapp, T.: A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. IEEE Trans. Signal Process. 50 (2002)
Chang, C.-C., Lin, C.-J.: LIBSVM: A Library for Support Vector Machines (2001). Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm
Freedman, J.L.: Crowding and Behavior (1975)
Grabner, H., Bischof, H.: On-line boosting and vision. In: CVPR (2006)
Hall, E.T.: The Hidden Dimension. Garden City (1966)
Helbing, D., Molnár, P.: Social force model for pedestrian dynamics. Phys. Rev. E 51(5), 4282–4286 (1995)
Lerner, A., Chrysanthou, Y., Lischinski, D.: Crowds by example. In: EUROGRAPHICS (2007)
Luber, M., Stork, J.A., Tipaldi, G.D., Arras, K.O.: People tracking with human motion predictions from social forces. In: 2010 IEEE International Conference on Robotics and Automation, pp. 464–469. IEEE, New York (May 2010)
Massive Software: Massive (2010)
McPhail, C., Wohlstein, R.T.: Using film to analyze pedestrian behavior. Sociol. Methods Res. 10(3), 347–375 (1982)
Mehran, R., Oyama, A., Shah, M.: Abnormal crowd behavior detection using Social Force model. In: CVPR (2009)
Pellegrini, S., Ess, A., Schindler, K., Gool, L.V.: You’ll never walk alone: Modeling social behavior for multi-target tracking. In: ICCV (2009)
Penn, A., Turner, A.: Space syntax based agent simulation. In: Pedestrian and Evacuation Dynamics (2002)
Reid, D.B.: An algorithm for tracking multiple targets. IEEE Trans. Autom. Control 24(6), 843–854 (1979)
Schadschneider, A.: Cellular automaton approach to pedestrian dynamics – theory. In: PED (2001)
Scovanner, P., Tappen, M.: Learning pedestrian dynamics from the real world. In: ICCV (2009)
Trautman, P., Krause, A.: Unfreezing the robot: Navigation in dense, interacting crowds. In: IROS (2010)
Ziebart, B.D., Ratliff, N., Gallagher, G., Mertz, C., Peterson, K., Andrew, J., Martial, B., Anind, H., Dey, K., Srinivasa, S.: Planning-based Prediction for Pedestrians (2009)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix A: Derivation of the marginal probabilities
Appendix A: Derivation of the marginal probabilities
In this appendix we are going to show a derivation for \(p(\mathbf{x}_{i}^{t},\mathbf{v}_{i}^{t})\), and \(p(\mathbf{x}_{i}^{t})\).
We start by factorizing the probability of the state S t
Now, we can assume that \(p(\mathbf{x}_{i}^{0},\mathbf{v}_{i}^{0})\) is initially given as a mixture of Gaussians (possibly with a single component). To show, by induction, that the marginal \(p(\mathbf {x}_{i}^{t},\mathbf{v}_{i}^{t})\) will have the form of a mixture of Gaussians, we assume that at time t−1 the distribution \(p(\mathbf{x}_{i}^{t-1},\mathbf{v}_{i}^{t-1})\) is already a mixture of Gaussians and prove that this is sufficient for \(p(\mathbf{x}_{i}^{t},\mathbf{v}_{i}^{t})\) to have the same form.
For the moment, the fact that the each factor of (24.19) is a mixture of Gaussians, say with H t component, means that P(S t) will be itself a mixture of Gaussians with M t=(H t)N components
where we use a mapping function φ:{1,…,H}N→{1,…,M} from each N-tuple in the Cartesian product of the h indices to a single m and we define \(w_{m}=w_{1h_{1}} w_{2h_{2}}\cdots w_{3h_{3}}\). \(\boldsymbol{\mu}_{ih}^{t}\) and \(\boldsymbol{\varSigma }_{ih}^{t}\) are the mean and the covariance matrix for the hth component of the mixture for subject i at time t, and \(\mu_{\mathbf{S}_{m}}^{t}\) and \(\boldsymbol{\varSigma }_{\mathbf{S}_{m}}^{t}\) are the mean and the covariance matrix for the joint state S t. Note that \(\boldsymbol{\mu}_{\mathbf{S}_{m}}^{t}\) is obtained by simply concatenating the \(\boldsymbol{\mu}_{ih_{i}}^{t}\) for each subject, and the covariance matrix \(\boldsymbol{\varSigma }_{\mathbf{S}_{m}}^{t}\) is a block diagonal matrix with blocks \(\boldsymbol{\varSigma }_{ih_{i}}^{t}\). Also, note that the φ mapping describes the possible world models for the set of subjects, as in each component of the mixture in (24.21) there is a single component h i selected from the mixture \(p(\mathbf{x}_{i}^{t},\mathbf{v}_{i}^{t})\).
Now let us see how to derive a single marginal \(p(\mathbf {x}_{i}^{t},\mathbf{v}_{i}^{t})\):
to meet real-time requirements, here we simplify the mixture of Gaussians in (24.21) by substituting each normal distribution with a Dirac function:
so that we can rewrite the integral in (24.22) as
By this approximation, we retain the mean and weight of the mixture components, but we discard the covariances. We will compensate for this in the empirical covariance that we will introduce later on. Continuing with the derivation we have
where
and where we define w mk =w m w k . So we show that \(p(\mathbf {x}_{i}^{t},\mathbf{v}_{i}^{t})\) has a mixture of Gaussian form with H t=M t−1 K components. As we noted above, because of the approximation in (24.23), we are discarding the uncertainty information included in the covariance matrix \(\boldsymbol{\varSigma }_{\mathbf{S}}^{t-1}\). We can partly compensate for this by appropriately modifying the covariance \(\boldsymbol{\varSigma }_{mk}^{t}\). In particular, we modify (24.30) by adding the position covariance at the previous time step as
Note that the velocity components of the covariance are in fact discarded, but we partially account for this in the empirical setting of Γ.
Finally, let us derive \(p(\mathbf{x}_{i}^{t})\) as
where we made use of the marginalization property of the Gaussian distribution.
Rights and permissions
Copyright information
© 2011 Springer-Verlag London Limited
About this chapter
Cite this chapter
Pellegrini, S., Ess, A., Van Gool, L. (2011). Predicting Pedestrian Trajectories. In: Moeslund, T., Hilton, A., Krüger, V., Sigal, L. (eds) Visual Analysis of Humans. Springer, London. https://doi.org/10.1007/978-0-85729-997-0_24
Download citation
DOI: https://doi.org/10.1007/978-0-85729-997-0_24
Publisher Name: Springer, London
Print ISBN: 978-0-85729-996-3
Online ISBN: 978-0-85729-997-0
eBook Packages: Computer ScienceComputer Science (R0)