Bayesian human segmentation in crowded situations

By Tao Zhao, Ramkant Nevatia. 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings., 2003

368 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/5ccc8b513c81ce22d35d6d5f011b4347d4751c48

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

Segmentation and Tracking of Multiple Humans in Crowded Environments (2008, methodology): “We have used prior on the number of objects in [53] to constrain oversegmentation.”
Privacy preserving crowd monitoring: Counting people without people models or tracking (2008, positive): “This is a canonical example of a problem that vision technology addresses with privacy invasive methods: detect the people in the scene [3, 4, 5, 6, 7], track them over time [8, 9, 10], and count the number of tracks.”
Bayesian Loss for Crowd Count Estimation With Point Supervision (2019, background): “In the past decade, a number of crowd counting algorithms [22, 58, 21, 12, 5, 35, 20, 11, 7, 31] have been proposed in the literature.”
Human detection from images and videos: A survey (2016, methodology): “Examples of this method are the works in [168, 54, 83, 183, 131].”
Measure Locally, Reason Globally: Occlusion-sensitive Articulated Pose Estimation (2006, background): “Drawing inspiration from [3] and [23], we define local likelihoods in terms of the product of individual pixel likelihoods in sub-regions of the image that are defined by the local state Xi.”
Detecting Pedestrians by Learning Shapelet Features (2007, background): “Zhao and Nevatia [21] work on difﬁcult scenes involving large crowds of people, and segment foreground blobs into individual people using a Markov chain Monte Carlo technique.”
Floor Fields for Tracking in High Density Crowd Scenes (2008, methodology): “The initial detection of objects was based on the their segmentation scheme ([13]) proposed earlier.”
Learning To Count Objects in Images (2010, background): “More generative approaches avoid nonmaximum suppression by reasoning about relations between object parts and instances [6, 14, 20, 33, 34], but they are still geared towards a situation with a small number of objects in images and require time-consuming inference.”
Crowd Analysis Using Computer Vision Techniques (2010, background): “Zhao and Nevatia [31] proposed a Bayesian approach to segment people in crowds.”
Optimized Pedestrian Detection for Multiple and Occluded People (2013, methodology): “One line of work proposes to segment foreground blobs into human shapes using an MCMC-based optimization approach to determine the number and configuration of overlapping shapes [23, 24, 10].”
A system for learning statistical motion patterns (2006, background): “While there exist some algorithms [23], [24], [28], [31], [39], [51] for tracking multiple objects, most of them fail due to the complexity of motions in crowded traffic scenes.”
Crowd Counting Using Multiple Local Features (2009, background): “For example, head detection has been proposed to estimate crowd sizes [10].”
Counting People With Low-Level Features and Bayesian Regression (2012, neutral): “Pedestrian detection algorithms can be based on boosting appearance and motion features [1], Bayesian model-based segmentation [2], [3], histogram-of-gradients [25], or integrated top-down and bottom-up processing [4].”
Dynamic Human Pose Estimation using Markov Chain Monte Carlo Approach (2005, methodology): “This technique is motivated by the data-driven MCMC framework [15][16].”
Counting Crowded Moving Objects (2006, background): “The dataset from [28]: an elevated view of a crowd consisting of zero to twelve persons.”
Detection of multiple, partially occluded humans in a single image by Bayesian combination of edgelet part detectors (2005, background): “Considerable progress has been made in detection and tracking of humans in video sequences, e.g. see [16].”
A Lagrangian Particle Dynamics Approach for Crowd Flow Segmentation and Stability Analysis (2007, methodology): “Representations based on xt slices of spatio-temporal video volume [7], shape and color models of individuals [ 9 ], boundary contours [3], and trajectories of interest points [5] have been used for this purpose.”