Human Action Recognition Using Adaptive Local Motion Descriptor in Spark

Uddin, Azher; Joolee, Joolekha Bibi; Alam, Aftab; Lee, Young-Koo

doi:10.1109/access.2017.2759225

Cited by 38 publications

(31 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed framework is able to process both the real-time video streams and batch video analytics. For the offline video processing, in the beginning, all the batch video data are stored in the Hadoop Distributed File System (HDFS) [8] and processed on the top of Spark [9,10]. In contrast, for online video processing, we adopted Apache Kafka [11] and Spark Streaming [12].…”

Section: Introductionmentioning

confidence: 99%

“…Besides, in References [14,15], the authors introduced a Streaming Video Engine (SVE) for uploading and processing videos in a distributed manner and a framework for real-time video analysis on Spark, respectively. Despite that, existing literature lacks support for the dynamic feature extraction on a distributed environment, which is a very essential component to provide any high-level services [10,16]. To address this issue, we introduce a distributed video processing library that provides video processing on top of Spark.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SIAT: A Distributed Video Analytics Framework for Intelligent Video Surveillance

Uddin

Alam

et al. 2019

Symmetry

Self Cite

View full text Add to dashboard Cite

In recent years, the amount of intelligent CCTV cameras installed in public places for surveillance has increased enormously and as a result, a large amount of video data is produced every moment. Due to this situation, there is an increasing request for the distributed processing of large-scale video data. In an intelligent video analytics platform, a submitted unstructured video undergoes through several multidisciplinary algorithms with the aim of extracting insights and making them searchable and understandable for both human and machine. Video analytics have applications ranging from surveillance to video content management. In this context, various industrial and scholarly solutions exist. However, most of the existing solutions rely on a traditional client/server framework to perform face and object recognition while lacking the support for more complex application scenarios. Furthermore, these frameworks are rarely handled in a scalable manner using distributed computing. Besides, existing works do not provide any support for low-level distributed video processing APIs (Application Programming Interfaces). They also failed to address a complete service-oriented ecosystem to meet the growing demands of consumers, researchers and developers. In order to overcome these issues, in this paper, we propose a distributed video analytics framework for intelligent video surveillance known as SIAT. The proposed framework is able to process both the real-time video streams and batch video analytics. Each real-time stream also corresponds to batch processing data. Hence, this work correlates with the symmetry concept. Furthermore, we introduce a distributed video processing library on top of Spark. SIAT exploits state-of-the-art distributed computing technologies with the aim to ensure scalability, effectiveness and fault-tolerance. Lastly, we implant and evaluate our proposed framework with the goal to authenticate our claims.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

SIAT: A Distributed Video Analytics Framework for Intelligent Video Surveillance

Uddin

Alam

et al. 2019

Symmetry

Self Cite

View full text Add to dashboard Cite

show abstract

“…A novel human activity recognition method is proposed, utilizing independent components of activity to mold information from image sequences and Hidden Markov Model (HMM) [22]. Similarly, Uddin et al [9] came up with a novel feature descriptor called Adaptive Local Motion Descriptor (ALMD) that considers motion and appearance and they utilize this feature descriptor to do batch processing in a Spark cluster environment. ALMD is a kind of combination of LBP combining two continuous frames for moving motion feature extraction.…”

Section: Related Workmentioning

confidence: 99%

“…However, the basic LBP has several drawbacks. It is susceptible to noise, as slight variation in the intensities of the neighbors can completely change the resultant binary code [9].…”

Section: Local Binary Pattern (Lbp)mentioning

confidence: 99%

“…Spark provides a comprehensive framework to manage and process big data. Uddin et al [9] performed action recognition using an ALMD (Adaptive Local Motion Descriptor) based feature descriptor and the Spark MLlib Random Forest [10]. Human action recognition using Distributed Trajectory based features and Distributed Gaussian Mixture Models is proposed in [11], while our work focuses on estimating the similarity between the user performed action video and reference action video that is stored on the database.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Similarity Estimation for Large-Scale Human Action Video Data on Spark

Uddin

Dolgorsuren

et al. 2018

Applied Sciences

Self Cite

View full text Add to dashboard Cite

Abstract:The amount of human action video data is increasing rapidly due to the growth of multimedia data, which increases the problem of how to process the large number of human action videos efficiently. Therefore, we devise a novel approach for human action similarity estimation in the distributed environment. The efficiency of human action similarity estimation depends on feature descriptors. Existing feature descriptors such as Local Binary Pattern and Local Ternary Pattern can only extract texture information but cannot obtain the object shape information. To resolve this, we introduce a new feature descriptor, namely Edge based Local Pattern descriptor (ELP). ELP can extract object shape information besides texture information and ELP can also deal with intensity fluctuations. Moreover, we explore Apache Spark to perform feature extraction in the distributed environment. Finally, we present an empirical scalability evaluation of the task of extracting features from video datasets.

show abstract

Analysis of Commuting Characteristics of Mobile Signaling Big Data Based on Spark

Suo

Lin

2018

Proceedings of the Fifth Euro-China Conference on Intelligent Data Analysis and Applications

View full text Add to dashboard Cite

Human Action Recognition Using Adaptive Local Motion Descriptor in Spark

Cited by 38 publications

References 29 publications

SIAT: A Distributed Video Analytics Framework for Intelligent Video Surveillance

SIAT: A Distributed Video Analytics Framework for Intelligent Video Surveillance

Similarity Estimation for Large-Scale Human Action Video Data on Spark

Analysis of Commuting Characteristics of Mobile Signaling Big Data Based on Spark

Contact Info

Product

Resources

About