M2Tracker: A Multi-View Approach to Segmenting and Tracking People in a Cluttered Scene Using Region-Based Stereo

Mittal, Anurag; Davis, Larry S.

doi:10.1007/3-540-47969-4_2

Cited by 252 publications

(260 citation statements)

References 2 publications

Supporting

Mentioning

258

Contrasting

Unclassified

Order By: Relevance

“…Kettnaker and Zabih [14] combined visual appearance matching with mutual content constraints between cameras to identify a same person from different cameras. M2 Tracker system [15] could segment and track people in cluttered environments by using region-based stereo from up to 16 cameras. However, multiple-camera solution is limited to applications of small spatial areas.…”

Section: Related Workmentioning

confidence: 99%

“…In surveillance, oblique settings of cameras are preferable due to a wider field of view. Others need a significant amount of work for settings and initialization when camera viewpoints change [4], [5], [13]- [15]. In practice, camera-viewpoint changing during the operation likely occurs due to both human and natural factors, such as carelessness in cleaning and maintaining cameras, and earthquakes, etc.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Real-Time Counting People in Crowded Areas by Using Local Empirical Templates and Density Ratios

Hung

Hsu

Chung

et al. 2012

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIn this paper, a fast and automated method of counting pedestrians in crowded areas is proposed along with three contributions. We firstly propose Local Empirical Templates (LET), which are able to outline the foregrounds, typically made by single pedestrians in a scene. LET are extracted by clustering foregrounds of single pedestrians with similar features in silhouettes. This process is done automatically for unknown scenes. Secondly, comparing the size of group foreground made by a group of pedestrians to that of appropriate LET captured in the same image patch with the group foreground produces the density ratio. Because of the local scale normalization between sizes, the density ratio appears to have a bound closely related to the number of pedestrians who induce the group foreground. Finally, to extract the bounds of density ratios for groups of different number of pedestrians, we propose a 3D human models based simulation in which camera viewpoints and pedestrians' proximity are easily manipulated. We collect hundreds of typical occluded-people patterns with distinct degrees of human proximity and under a variety of camera viewpoints. Distributions of density ratios with respect to the number of pedestrians are built based on the computed density ratios of these patterns for extracting density ratio bounds. The simulation is performed in the offline learning phase to extract the bounds from the distributions, which are used to count pedestrians in online settings. We reveal that the bounds seem to be invariant to camera viewpoints and humans' proximity. The performance of our proposed method is evaluated with our collected videos and PETS 2009's datasets. For our collected videos with the resolution of 320x240, our method runs in real-time with good accuracy and frame rate of around 30 fps, and consumes a small amount of computing resources. For PETS 2009's datasets, our proposed method achieves competitive results with other methods tested on the same datasets [1], [2]. key words: local empirical templates, local density ratios, density ratio bounds, and people counting

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Real-Time Counting People in Crowded Areas by Using Local Empirical Templates and Density Ratios

Hung

Hsu

Chung

et al. 2012

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…We confine our discussion to multiple cameras with overlapping viewpoints, for we are primarily interested in resolving ambiguities due to occlusion. Mittal and Davis [13] match color regions from all pairs of camera views to generate a ground plane occupancy map by kernel density estimation. In Khan et.al.…”

Section: Related Workmentioning

confidence: 99%

Crowd Detection with a Multiview Sampler

Collins

2010

Computer Vision – ECCV 2010

View full text Add to dashboard Cite

Abstract. We present a Bayesian approach for simultaneously estimating the number of people in a crowd and their spatial locations by sampling from a posterior distribution over crowd configurations. Although this framework can be naturally extended from single to multiview detection, we show that the naive extension leads to an inefficient sampler that is easily trapped in local modes. We therefore develop a set of novel proposals that leverage multiview geometry to propose global moves that jump more efficiently between modes of the posterior distribution. We also develop a statistical model of crowd configurations that can handle dependencies among people and while not requiring discretization of their spatial locations. We quantitatively evaluate our algorithm on a publicly available benchmark dataset with different crowd densities and environmental conditions, and show that our approach outperforms other state-of-the-art methods for detecting and counting people in crowds.

show abstract

“…However this method is able to track a single target which restricts its application. Stereo vision techniques combined with shape matching [5,12], color information matching [10,17] or probabilistic models [17,6] can determine whether objects observed in different views are the same. After detecting the different objects in the scene, their counting and tracking become easy to accomplish.…”

Section: Introductionmentioning

confidence: 99%

People detection and tracking using the Explorative Particle Filtering

Saboune

Laganière

2009

2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops

View full text Add to dashboard Cite

Automatic people detection and tracking is a very essential task of video surveillance systems. It can improve a system's performance in important fields such as security, safety, human activity monitoring etc. In this paper we present a novel approach for people detection and 3D tracking. Our method is based on a human upper body 3D model and a likelihood function to evaluate its presence in a certain region of the scene. We then find the maxima of this function using a modified particle filtering algorithm which we call Explorative Particle Filtering (ExPF). We designed this algorithm in a way to guarantee a multiple objects tracking and a good estimation of their positions when using a small number of particles. Our technique is generic and simple as no dynamic models nor trained features models (color, shape etc.) were used. We also show some tracking results from video surveillance feeds in order to illustrate our approach.

show abstract

M2Tracker: A Multi-View Approach to Segmenting and Tracking People in a Cluttered Scene Using Region-Based Stereo

Cited by 252 publications

References 2 publications

Real-Time Counting People in Crowded Areas by Using Local Empirical Templates and Density Ratios

Real-Time Counting People in Crowded Areas by Using Local Empirical Templates and Density Ratios

Crowd Detection with a Multiview Sampler

People detection and tracking using the Explorative Particle Filtering

Contact Info

Product

Resources

About