Learning motion rules from real data: Neural network for crowd simulation

Abstract This paper addresses the problem of efficiently simulating a believable virtual crowd. Our method is the first one that uses the Neural Network (NN) model to fit behaviors from real crowd data to a crowd simulation. Unlike several rule-based approaches that often result in ‘walking robots’, our model can learn motion rules derived from real data and later simulate human walking motions. Additionally, unlike the existing data-driven crowd simulation methods that have to perform search operations on the bound dataset simultaneously during the simulation, our model directly uses the NN model to generate the proper motion for each crowd member. The proposed method is being tested on various scenarios and compared with state-of-the-art state-action-based methods that are commonly employed in data-driven crowd simulation systems. The results demonstrate a significant increase in speed, as well as better simulation quality.

[1]  H. Guterman,et al.  Knowledge extraction from artificial neural network models , 1997, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation.

[2]  Ding Yuan,et al.  ARCH: Adaptive recurrent-convolutional hybrid networks for long-term action recognition , 2016, Neurocomputing.

[3]  Dinesh Manocha,et al.  A statistical similarity measure for aggregate crowd dynamics , 2012, ACM Trans. Graph..

[4]  Dong Wang,et al.  A Gaussian mixture framework for incremental nonparametric regression with topology learning neural networks , 2016, Neurocomputing.

[5]  Dinesh Manocha,et al.  Directing Crowd Simulations Using Navigation Fields , 2011, IEEE Transactions on Visualization and Computer Graphics.

[6]  Jorge J. Moré,et al.  The Levenberg-Marquardt algo-rithm: Implementation and theory , 1977 .

[7]  Norman I. Badler,et al.  Context-sensitive data-driven crowd simulation , 2013, VRCAI '13.

[8]  P. Ramasamy,et al.  Wind speed prediction in the mountainous region of India using an artificial neural network model , 2015 .

[9]  Norman I. Badler,et al.  Generating a multiplicity of policies for agent steering in crowd simulation , 2015, Comput. Animat. Virtual Worlds.

[10]  Xiaogang Wang,et al.  Learning Collective Crowd Behaviors with Dynamic Pedestrian-Agents , 2014, International Journal of Computer Vision.

[11]  S. Karsoliya,et al.  Approximating Number of Hidden layer neurons in Multiple Hidden Layer BPNN Architecture , 2012 .

[12]  Dinesh Manocha,et al.  Reciprocal n-Body Collision Avoidance , 2011, ISRR.

[13]  Wentong Cai,et al.  Learning Behavior Patterns from Video: A Data-driven Framework for Agent-based Crowd Modeling , 2015, AAMAS.

[14]  Abhijit S. Pandya,et al.  Pattern Recognition with Neural Networks in C++ , 1995 .

[15]  Xuelong Li,et al.  Measuring Collectiveness via Refined Topological Similarity , 2016, ACM Trans. Multim. Comput. Commun. Appl..

[16]  Yiorgos Chrysanthou,et al.  The PAG Crowd: A Graph Based Approach for Efficient Data‐Driven Crowd Simulation , 2014, Comput. Graph. Forum.

[17]  Guang-Bin Huang,et al.  Learning capability and storage capacity of two-hidden-layer feedforward networks , 2003, IEEE Trans. Neural Networks.

[18]  Stephen John Turner,et al.  Evaluation of Crowd Models in Low Density Scenarios Using Real-World Crowd Data , 2015, 2015 IEEE/ACM 19th International Symposium on Distributed Simulation and Real Time Applications (DS-RT).

[19]  Andrzej Szopa,et al.  Postural pattern recognition in children with unilateral cerebral palsy , 2014, Therapeutics and clinical risk management.

[20]  Geoff S. Nitschke,et al.  Controlling Crowd Simulations using Neuro-Evolution , 2015, GECCO.

[21]  Petros Faloutsos,et al.  SteerFit: automated parameter fitting for steering algorithms , 2014, SCA '14.

[22]  Sayan Ghosh,et al.  Estimation of ash, moisture content and detection of coal lithofacies from well logs using regression and artificial neural network modelling , 2016 .

[23]  Dimitris N. Metaxas,et al.  Eurographics/ Acm Siggraph Symposium on Computer Animation (2007) Group Behavior from Video: a Data-driven Approach to Crowd Simulation , 2022 .

[24]  Kok Kiong Tan,et al.  An RBF neural network approach towards precision motion system with selective sensor fusion , 2016, Neurocomputing.

[25]  Jaroslaw Was,et al.  Towards realistic and effective Agent-based models of crowd dynamics , 2014, Neurocomputing.

[26]  Maryam Mosleh,et al.  Simulation and evaluation of interval-valued fuzzy linear Fredholm integral equations with interval-valued fuzzy neural network , 2016, Neurocomputing.

[27]  Dong Wang,et al.  Anomaly Detection in Traffic Scenes via Spatial-Aware Motion Reconstruction , 2017, IEEE Transactions on Intelligent Transportation Systems.

[28]  P. Torrens,et al.  Building Agent‐Based Walking Models by Machine‐Learning on Diverse Databases of Space‐Time Trajectory Samples , 2011 .

[29]  Lucas Kovar,et al.  Motion graphs , 2002, SIGGRAPH Classes.

[30]  Okyay Kaynak,et al.  Control of a direct drive robot using fuzzy spiking neural networks with variable structure systems-based learning algorithm , 2015, Neurocomputing.

[31]  Stephen John Turner,et al.  A Data-Driven Crowd Simulation Model Based on Clustering and Classification , 2013, 2013 IEEE/ACM 17th International Symposium on Distributed Simulation and Real Time Applications.

[32]  Vladimir Pavlovic,et al.  Filling in the blanks: reconstructing microscopic crowd motion from multiple disparate noisy sensors , 2016, 2016 IEEE Winter Applications of Computer Vision Workshops (WACVW).