Privacy preserving crowd monitoring: Counting people without people models or tracking

Chan, Antoni B.; Liang, Z.-S.J.; Vasconcelos, Nuno

doi:10.1109/cvpr.2008.4587569

Cited by 1,012 publications

(864 citation statements)

References 22 publications

Supporting

Mentioning

818

Contrasting

Unclassified

Order By: Relevance

“…By performing the count on features, in the case of cells, a single cell is not identified with these kind of methods, but by looking at the higher level features like color, edges, etc., these methods give a direct estimate of the count. In the case of crowds [7], this has privacy advantages because their method does not directly recognise a single individual, which will give us privacy sensitive information. Often machine learning methods for finding or identifying a single individual object are available, but the feature level approach in this case does not use this information.…”

Section: Related Workmentioning

confidence: 99%

“…In [11,17], the estimation of automatic land cover categories is improved based on the confusion matrix, where these papers use a confusion matrix (decision-level) determined from groundtruth to correct the under and overestimates. In computer vision, counting cells [19] and crowds [7] is often performed using regression on the image features, which achieve very accurate counts (feature level). By performing the count on features, in the case of cells, a single cell is not identified with these kind of methods, but by looking at the higher level features like color, edges, etc., these methods give a direct estimate of the count.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Uncertainty-aware estimation of population abundance using machine learning

et al. 2015

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Uncertainty-aware estimation of population abundance using machine learning

et al. 2015

View full text Add to dashboard Cite

“…Some recent methods (e.g. Kong et al 2006;Chan et al 2008) attack this problem by reducing the objective from localization to counting, i.e. given an input image, estimate the number of present objects without explicitly telling their locations and sizes.…”

Section: Detection Of Multiple Occluded Objectsmentioning

confidence: 99%

Detection and Segmentation of Multiple, Partially Occluded Objects by Grouping, Merging, Assigning Part Detection Responses

Nevatia

2008

Int J Comput Vis

135

View full text Add to dashboard Cite

We propose a method that detects and segments multiple, partially occluded objects in images. A part hierarchy is defined for the object class. Both the segmentation and detection tasks are formulated as binary classification problem. A whole-object segmentor and several part detectors are learned by boosting local shape feature based weak classifiers. Given a new image, the part detectors are applied to obtain a number of part responses. All the edge pixels in the image that positively contribute to the part responses are extracted. A joint likelihood of multiple objects is defined based on the part detection responses and the object edges. Computation of the joint likelihood includes an inter-object occlusion reasoning that is based on the object silhouettes extracted with the whole-object segmentor. By maximizing the joint likelihood, part detection responses are grouped, merged, and assigned to multiple object hypotheses. The proposed approach is demonstrated with the class of pedestrians. The experimental results show that our method outperforms the previous ones.

show abstract

“…During learning, an observed set of feature descriptors, such as edges and texture, is correlated with the number of people present. During operation, the feature descriptors are classified and the estimated number of people in the scene is obtained as a result, preserving privacy to some extent [1,12].…”

Section: Related Workmentioning

confidence: 99%

Counting Pedestrians in Bidirectional Scenarios Using Zenithal Depth Images

Vera

Zenteno

Salas

2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this document, we describe a people counting system that can precisely detect people as they are seen from a zenithal depth camera pointing at the floor. In particular, we are interested in scenarios where there are two preferred directions of motion. In our framework, we detect people using a Support Vector Machine classifier, follow their trajectory by modeling the problem of matching observations between frames as a bipartite graph, and determine the direction of their motion with a bi-directional classifier. We include experimental evidence, from four different scenarios, for each major stage of our method.

show abstract

Privacy preserving crowd monitoring: Counting people without people models or tracking

Abstract: We present a privacy-preserving

Cited by 1,012 publications

References 22 publications

Uncertainty-aware estimation of population abundance using machine learning

Uncertainty-aware estimation of population abundance using machine learning

Detection and Segmentation of Multiple, Partially Occluded Objects by Grouping, Merging, Assigning Part Detection Responses

Counting Pedestrians in Bidirectional Scenarios Using Zenithal Depth Images

Contact Info

Product

Resources

About