Unsupervised Bayesian Detection of Independent Motion in Crowds

By Gabriel J. Brostow, Roberto Cipolla. Computer Vision and Pattern Recognition, 2006

455 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/db2e05ccd1d6d106ee8816bf65c655f250f7206c

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 Computer Vision and Pattern Recognition

Citations in the corpus

Feature Mining for Localised Crowd Counting (2012, background): “In counting by clustering [4, 20], a crowd is assumed to be composed of individual entities, each of which has unique yet coherent motion patterns”
Abnormal crowd behavior detection using social force model (2009, background): “Therefore, to understand the crowd behavior it is necessary to perform segmentation or detect objects to analyze group behaviors [7].”
Crowd Counting via Adversarial Cross-Scale Consistency Pursuit (2018, methodology): “Methods [1, 6, 5, 12, 2, 22, 4] use regressors trained with low-level features (e.g. HOG, SIFT, Fourier Analysis, detections and trajectories) to predict global counts.”
You'll never walk alone: Modeling social behavior for multi-target tracking (2009, background): “Others have developed approaches specifically for crowded scenes [1, 6, 24].”
Video Segmentation with Superpixels (2012, unknown): “Recent works on VS have employed point trajectories [2], improving on the corresponding point track clustering literature [24, 25], and dense solutions are obtained with densification by non-linear diffusion [23] and graph-based methods [16].”
Crowd Analysis Using Computer Vision Techniques (2010, methodology): “Brostow and Cipolla [35] presented an unsupervised Bayesian clustering method to detect independent movements in a crowd.”
Multi-source Multi-scale Counting in Extremely Dense Crowd Images (2013, background): “Brostow and Cipolla [3] and Rabaud and Belongie [19] count moving objects by estimating contiguous regions of coherent motion.”
Vision-Based Analysis of Small Groups in Pedestrian Crowds (2012, background): “A number of works have considered the problem of counting pedestrians by clustering corner feature trajectories into groups that each represent a single individual [68]–[70].”
Switching Convolutional Neural Network for Crowd Counting (2017, methodology): “In crowd counting from videos, [3] use image features like Tomasi-Kanade features into a motion clustering framework.”
Unsupervised Detection and Tracking of Arbitrary Objects with Dependent Dirichlet Process Mixtures (2012, methodology): “Other attempts to perform unsupervised detection and tracking include methods for clustering short sequences of positions extracted by detecting the motion of objects [8, 9], which aim to return full-length distinct object tracks, and graph based methods that carry out a similar task using spectral ”
Tracking Multiple Occluding People by Localizing on Multiple Scene Planes (2009, methodology): “For a detailed review of the state of the art in tracking research, the reader is referred to the recent survey by Yilmaz et al. [61].”
Multiple object tracking: A literature review (2014, methodology): “As long as some parts of an object are visible, the clustering of feature point trajectories will work [62], [68], [112].”
Context-Aware Crowd Counting (2018, background): “Even though approaches relying on low-level features [9, 6, 4, 27, 7, 14] can yield good results, they have now mostly been superseded by CNN-based methods [42, 31, 5], a survey of which can be found in [36].”
Cross-scene crowd counting via deep convolutional neural networks (2015, background): “Many works have been proposed to count the pedestrians by detection [29, 14, 27] or trajectory-clustering [3, 21].”
Crowd analysis: a survey (2008, background): “[17] tackle the problem by tracking simple image features and probabilistically grouping them into clusters representing independently moving entities.”
A Survey of Recent Advances in CNN-based Single Image Crowd Counting and Density Estimation (2017, background): “Various approaches have been proposed to tackle the problem of crowd counting in images [41, 19, 52, 107, 111] and videos [12, 35, 77, 21].”
Floor Fields for Tracking in High Density Crowd Scenes (2008, background): “[12] presented a probabilistic framework for the clustering of feature point trajectories for detection and tracking of individual pedestrians in crowds.”