Spatio-Temporal Optical Flow Analysis for People Counting

Benabbas, Yassine; Ihaddadene, Nacim; Yahiaoui, Tarek; Urruty, Thierry; Djeraba, Chabane

doi:10.1109/avss.2010.29

Cited by 24 publications

(9 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The counting by regression approach, which we focus on in this work, aims at mapping from lowlevel image features to the number of people or to the density map of a scene by supervised training of a regression model. The earliest methods estimate the number of people using holistic scene descriptors such as foreground segment (Ma et al, 2004), edge (Kong et al, 2005), texture and gradient (Wu et al, 2006;Ojala et al, 2002), shape (Dong et al, 2007), intensity (Lempitsky and Zisserman, 2010) and motion (Benabbas et al, 2010). The main regression models used in such methods are Linear Regression, Random Forests and Support Vector Regression (Loy et al, 2013).…”

Section: Related Workmentioning

confidence: 99%

Investigating Synthetic Data Sets for Crowd Counting in Cross-scene Scenarios

Delussu

Putzu

Fumera

2020

Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

Crowd counting and density estimation are crucial functionalities in intelligent video surveillance systems but are also very challenging computer vision tasks in scenarios characterised by dense crowds, due to scale and perspective variations, overlapping and occlusions. Regression-based crowd counting models are used for dense crowd scenes, where pedestrian detection is infeasible. We focus on real-world, cross-scene application scenarios where no manually annotated images of the target scene are available for training regression models, but only images with different backgrounds and camera views can be used (e.g., from publicly available data sets), which can lead to low accuracy. To overcome this issue, we propose to build the training set using synthetic images of the target scene, which can be automatically annotated with no manual effort. This work provides a preliminary empirical evaluation of the effectiveness of the above solution. To this aim, we carry out experiments using real data sets as the target scenes (testing set) and using different kinds of synthetically generated crowd images of the target scenes as training data. Our results show that synthetic training images can be effective, provided that also their background, beside their perspective, closely reproduces the one of the target scene.

show abstract

Section: Related Workmentioning

confidence: 99%

Investigating Synthetic Data Sets for Crowd Counting in Cross-scene Scenarios

Delussu

Putzu

Fumera

2020

Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

show abstract

“…In general, linear model is expected to give poorer performance as its linear property imposes a limitation on the model in capturing only the linear relationship between the people count and low-level features [4,22,46]. In most cases especially in crowded environments, the visual observations and people count will not be linearly related.…”

Section: Findings and Analysismentioning

confidence: 99%

“…In a sparse scene where smaller crowd size and fewer inter-object occlusions are observed, the aforementioned linear regressor [4,22,46] may suffice since the mapping between the observations and people count typically presents a linear relationship. Nevertheless, given a more crowded environment with severe inter-object occlusion, one may have to employ a nonlinear regressor to adequately capture the nonlinear trend in the feature space [9].…”

Section: Regression Modelsmentioning

confidence: 99%

“…To achieve this, an EM approach similar to [4] is used. The patch to hypothesis assignments is defined by an assignment vector V = {vi} of length N with vi 2 [1, .…”

Section: Counting By Clusteringmentioning

confidence: 99%

“…• Least-squares support vector regression (LSSVR) -linear kernel with four-fold cross-validation for parameter optimisation • Gaussian processes regression (GPR) -linear kernel + RBF kernel as in [13] 4 .…”

Section: • Linear Regression (Lr)mentioning

confidence: 99%

See 2 more Smart Citations

Crowd Counting and Profiling: Methodology and Evaluation

Loy

Chen

Gong

et al. 2013

The International Series in Video Computing

175

View full text Add to dashboard Cite

Video imagery based crowd analysis for population profiling and density estimation in public spaces can be a highly effective tool for establishing global situational awareness. Different strategies such as counting by detection and counting by clustering have been proposed, and more recently counting by regression has also gained considerable interest due to its feasibility in handling relatively more crowded environments. However, the scenarios studied by existing regression-based techniques are rather diverse in terms of both evaluation data and experimental settings. It can be difficult to compare them in order to draw general conclusions on their effectiveness. In addition, contributions of individual components in the processing pipeline such as feature extraction and perspective normalisation remain unclear and less well studied. This study describes and compares the state-of-the-art methods for video imagery based crowd counting, and provides a systematic evaluation of different methods using the same protocol. Moreover, we evaluate critically each processing component to identify potential bottlenecks encountered by existing techniques. Extensive evaluation is conducted on three public scene datasets, including a new shopping centre environment with labelled ground truth for validation. Our study reveals new insights into solving the problem of crowd analysis for population profiling and density estimation, and considers open questions for future studies.

show abstract