Unsupervised Universal Attribute Modeling for Action Recognition

Roy, Debaditya; Murty, K. Sri Rama; Mohan, C. Krishna

doi:10.1109/tmm.2018.2887021

Cited by 57 publications

(21 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The best results of RGB-only models were R(2 + 1)D-RGB [53] and I3D-RGB [9], but they used more complex models with more params or FLOPs. We note that two-stream I3D [9] obtains 98.0% on UCF101 and 80.7% on HMDB51 top-1 accuracy, and some recent works on UCF101 [59] and HMDB51 [59], [60], [60]- [62] even achieve better results, but they are two-stream models with pre-computed optical flow, even more extra information such as audio signals.…”

Section: We Also Provide the Visualization Of Spatiotemporal Attentiomentioning

confidence: 77%

Learning Attention-Enhanced Spatiotemporal Representation for Action Recognition

et al. 2020

View full text Add to dashboard Cite

Learning spatiotemporal features via 3D-CNN (3D Convolutional Neural Network) models has been regarded as an effective approach for action recognition. In this paper, we explore visual attention mechanism for video analysis and propose a novel 3D-CNN model, dubbed AE-I3D (Attention-Enhanced Inflated-3D Network), for learning attention-enhanced spatiotemporal representation. The contribution of our AE-I3D is threefold: First, we inflate soft attention in spatiotemporal scope for 3D videos, and adopt softmax to generate probability distribution of attentional features in a feedforward 3D-CNN architecture; Second, we devise an AE-Res (Attention-Enhanced Residual learning) module, which learns attention-enhanced features in a two-branch residual learning way, also the AE-Res module is lightweight and flexible, so that can be easily embedded into many 3D-CNN architectures; Finally, we embed multiple AE-Res modules into an I3D (Inflated-3D) network, yielding our AE-I3D model, which can be trained in an end-to-end, video-level manner. Different from previous attention networks, our method inflates residual attention from 2D image to 3D video for 3D attention residual learning to enhance spatiotemporal representation. We use RGB-only video data for evaluation on three benchmarks: UCF101, HMDB51, and Kinetics. The experimental results demonstrate that our AE-I3D is effective with competitive performance.

show abstract

Section: We Also Provide the Visualization Of Spatiotemporal Attentiomentioning

confidence: 77%

Learning Attention-Enhanced Spatiotemporal Representation for Action Recognition

et al. 2020

View full text Add to dashboard Cite

show abstract

“…For the purpose of action recognition in surveillance scenes [57] proposes a Gaussian mixture model called Universal Attribute Modelling (UAM) using unsupervised learning approach. The UAM is also been used for facial expression recognition where it captures the attributes of all expressions [58] Further, for autonomous vehicles like cars or UAVs (Unmanned Aerial Vehicles) it is very essential to distinguish between normal and abnormal states.…”

Section: Unsupervised Learningmentioning

confidence: 99%

“…The overall framework should ensure that the nodes are not overloaded with work however if they are overloaded in the peak hours, the tasks should be partitioned and scheduled accordingly [106] [107]. Objects (Grassy Scene) AVIRIS [72] Objects (Aerial View) SkyEye [41] Objects (Aerial View) UCF101 [57] Individual (Action Videos) HMDB51 [57] Individual (Action Videos) THUMOS14 [57] [102] Individual (Action Videos) BP4D [58] Individual (Facial Expressions) AFEW [58] Individual (Facial Expressions) CAVIAR [67] Individuals, Objects (Vehicles) PASCAL VOC [6] [47] [69] [89] Objects (Vehicles) DS1, DS2 [8] Events (Fire/Smoke/Fog) SFpark [32] Object (Taxis) LISA 2010 [6] Objects (Vehicles) ImageNet [42] Events (Fire/Smoke/Fog) QMUL Junction [64] [49] Objects (Vehicles)…”

Section: Challenge 3: Quality Of Servicementioning

confidence: 99%

Anomaly Detection using Edge Computing in Video Surveillance System: Review

Patrikar¹,

Parate²

2021

Preprint

View full text Add to dashboard Cite

The current concept of Smart Cities influences urban planners and researchers to provide modern, secured and sustainable infrastructure and give a decent quality of life to its residents. To fulfill this need video surveillance cameras have been deployed to enhance the safety and well-being of the citizens. Despite technical developments in modern science, abnormal event detection in surveillance video systems is challenging and requires exhaustive human efforts. In this paper, we surveyed various methodologies developed to detect anomalies in intelligent video surveillance. Firstly, we revisit the surveys on anomaly detection in the last decade. We then present a systematic categorization of methodologies developed for ease of understanding. Considering the notion of anomaly depends on context, we identify different objects-of-interest and publicly available datasets in anomaly detection. Since anomaly detection is considered a time-critical application of computer vision, our emphasis is on anomaly detection using edge devices and approaches explicitly designed for them. Further, we discuss the challenges and opportunities involved in anomaly detection at the edge.

show abstract

“…Action recognition is an important aspect of video content analysis and has been extensively studied over the last few years [16], [17], [18], [19], [20], [21]. Earlier methods are mostly based on hand-crafted visual features [22], [23].…”

Section: A Action Recognitionmentioning

confidence: 99%

Temporal Action Localization using Long Short-Term Dependency

Zhou¹,

Li²,

Kung³

2019

Preprint

View full text Add to dashboard Cite

Temporal action localization in untrimmed videos is an important but difficult task. Difficulties are encountered in the application of existing methods when modeling temporal structures of videos. In the present study, we developed a novel method, referred to as Gemini Network, for effective modeling of temporal structures and achieving high-performance temporal action localization. The significant improvements afforded by the proposed method are attributable to three major factors. First, the developed network utilizes two subnets for effective modeling of temporal structures. Second, three parallel feature extraction pipelines are used to prevent interference between the extractions of different stage features. Third, the proposed method utilizes auxiliary supervision, with the auxiliary classifier losses affording additional constraints for improving the modeling capability of the network. As a demonstration of its effectiveness, the Gemini Network was used to achieve state-of-the-art temporal action localization performance on two challenging datasets, namely, THUMOS14 and ActivityNet.

show abstract

Unsupervised Universal Attribute Modeling for Action Recognition

Cited by 57 publications

References 28 publications

Learning Attention-Enhanced Spatiotemporal Representation for Action Recognition

Learning Attention-Enhanced Spatiotemporal Representation for Action Recognition

Anomaly Detection using Edge Computing in Video Surveillance System: Review

Temporal Action Localization using Long Short-Term Dependency

Contact Info

Product

Resources

About