Action recognition using cuboids of interest points

Vishwakarma, Sarvesh; Sapre, Akshay; Agrawal, Anupam

doi:10.1109/icspcc.2011.6061680

Cited by 12 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They used a longest common subsequence algorithm to verify different sets of trajectories. Vishwakarma and Agrawal [45] considered multiclass activities fused in a threedimensional (spatial and time) coordinate activity recognition system to achieve maximum accuracy. They quantized feature vectors of interest points utilizing a histogram.…”

Section: Introductionmentioning

confidence: 99%

A new method for violence detection in surveillance scenes

Zhang

Yang

Jia

et al. 2015

Multimed Tools Appl

112

View full text Add to dashboard Cite

Violence detection is a hot topic for surveillance systems. However, it has not been studied as much as for action recognition. Existing vision-based methods mainly concentrate on violence detection and make little effort to determine the location of violence. In this paper, we propose a fast and robust framework for detecting and localizing violence in surveillance scenes. For this purpose, a Gaussian Model of Optical Flow (GMOF) is proposed to extract candidate violence regions, which are adaptively modeled as a deviation from the normal behavior of crowd observed in the scene. Violence detection is then performed on each video volume constructed by densely sampling the candidate violence regions. To distinguish violent events from nonviolent events, we also propose a novel descriptor, named as Orientation Histogram of Optical Flow (OHOF), which are fed into a linear SVM for classification. Experimental results on several benchmark datasets have demonstrated the superiority of our proposed method over the state-of-the-arts in terms of both detection accuracy and processing speed, even in crowded scenes.

show abstract

Section: Introductionmentioning

confidence: 99%

A new method for violence detection in surveillance scenes

Zhang

Yang

Jia

et al. 2015

Multimed Tools Appl

112

View full text Add to dashboard Cite

show abstract

“…Flow Field Model: Generally, we need to know how crowds change with respect to times and try to discover some common designs. In this manner, we could anticipate crowd abnormal behaviors [18].…”

Section: The Physics Inspired Approachmentioning

confidence: 99%

Abnormal Crowd behavior Detection Using Structural Context Descriptor

Pooja¹,

Vishwakarma²

2016

BIJAIP

Self Cite

View full text Add to dashboard Cite

In concern with the people safety in public places, an Abnormal Behavior Detection in crowded scene has gained popularity. The Abnormal Behavior Detection is an active area of research. This has gained a particular interest amongst the researchers in this evolving era of computer science. For this purpose the system proposes the structure context descriptor (SCD) for describing the crowd individual, which represents the relationship among the individuals. The approach presents the abnormal behavior detection method from the crowd scenes to efficiently detect anomalies present in the crowded scenes.

show abstract

“…Oikonomopoulous et al [43] represented a human action as a collection of short trajectories extracted in the areas having significant amounts of visual activities in a scene. Vishwakarma and Agrawal [44] considered multiclass activities fused in a three dimensional (spatial and time) coordinate system to achieve maximum accuracy. This method worked well in semantically varying events and was robust to scale and view changes.…”

Section: A Related Work On Action Recognitionmentioning

confidence: 99%

Discriminative Dictionary Learning With Motion Weber Local Descriptor for Violence Detection

Zhang

Jia

et al. 2017

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Abstract-Automatic violence detection from video is a hot topic for many video surveillance applications. However, there has been little success in developing an algorithm that can detect violence in surveillance videos with high performance. In this paper, following our recently proposed idea of Motion Weber Local Descriptor (MoWLD), we make two major improvements and propose a more effective and efficient algorithm for detecting violence from motion images. First, we propose an improved WLD (IWLD) to better depict low-level image appearance information, and then extend the spatial descriptor IWLD by adding a temporal component to capture local motion information and hence form the MoIWLD. Second, we propose a modified sparse-representation-based classification (SRC) model to both control the reconstruction error of coding coefficients and minimize the classification error. Based on the proposed sparse model, a class-specific dictionary containing dictionary atoms corresponding to the class labels is learned by using class labels of training samples. With this learned dictionary, not only the representation residual but also the representation coefficients become discriminative. A classification scheme integrating the modified sparse model is developed to exploit such discriminative information. Experimental results on three benchmark datasets have demonstrated the superior performance of the proposed approach over the state-of-the-arts.

show abstract

Action recognition using cuboids of interest points

Cited by 12 publications

References 16 publications

A new method for violence detection in surveillance scenes

A new method for violence detection in surveillance scenes

Abnormal Crowd behavior Detection Using Structural Context Descriptor

Discriminative Dictionary Learning With Motion Weber Local Descriptor for Violence Detection

Contact Info

Product

Resources

About