A Lagrangian Particle Dynamics Approach for Crowd Flow Segmentation and Stability Analysis

By Saad Ali, Mubarak Shah. 2007 IEEE Conference on Computer Vision and Pattern Recognition, 2007

656 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Ocean Engineering, Signal Processing

View Paper: https://www.semanticscholar.org/paper/f6461ce6d9c12248227845da006e98c0bd4b33b0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Ocean Engineering

In Signal Processing

Citations in the corpus

Crowd Counting Using Multiple Local Features (2009, background): “crowd stability [1] (using optical flow over time).”
The Way They Move: Tracking Multiple Targets with Similar Appearance (2013, background): “Crowd is from the crowd UCF dataset [1].”
NWPU-Crowd: A Large-Scale Benchmark for Crowd Counting and Localization (2020, methodology): “From the category-wise results of Box Level, we find that all methods perform poorly for tiny heads (area range is [1, 10]).”
Crowded Scene Analysis: A Survey (2015, methodology): “Ali et al. [12] proposed the Lagrangian particle dynamic to segment high density crowd ﬂows.”
Crowd Analysis Using Computer Vision Techniques (2010, methodology): “This type of analysis is mostly used in two main applications: to measure the level of crowd comfort and to detect some specific events, as described in [6]–[8], e.g., unusual group behavior.”
Forecasting Interactive Dynamics of Pedestrians with Fictitious Play (2016, background): “Pioneering works have attempted to address this by parameterizing human behaviors with models such as social forces [13, 30], potential fields [2], and flow fields [3, 4].”
Layered Dynamic Textures (2005, methodology): “The authors thank Rene Vidal for the code from [11], [12], Mubarak Shah for the crowd videos [ 32 ], [57], Renaud Pe´teri, Mark Huiskes, and Sa´ndor Fazekas for the windmill video [58], and the anonymous reviewers for insightful comments.”
Anomaly detection in crowded scenes (2010, background): “Problems such as segmenting video into crowd components [3], estimating crowd size [17], determining the goal of individuals within a crowd [4] have all been subjects of research.”
Chaotic invariants of Lagrangian particle trajectories for anomaly detection in crowded scenes (2010, methodology): “Limiting our review of the literature to work directly related to analysis of crowded scenes, common features used for modeling crowded motions include optical flow [9, 10, 11, 12], gradient [13], spatiotemporal volume [14], etc.”
Abnormal crowd behavior detection using social force model (2009, methodology): “Similar to [2], we put a grid of particles over the image frame and move them with the flow field computed from the optical flow.”
Floor Fields for Tracking in High Density Crowd Scenes (2008, methodology): “Next, a grid of particles is overlaid on the first flow field of the volume and numerically advected [11].”