Bibliographic Summary of papers in different digital repositories
ScienceDirect
1 | SVAS: Surveillance Video Analysis System [1] |
2 | Jointly learning perceptually heterogeneous features for blind 3D video quality assessment [2] |
3 | Learning to detect video events from zero or very few video examples [3] |
4 | Learning an event-oriented and discriminative dictionary based on an adaptive label-consistent K-SVD method for event detection in soccer videos [4] |
5 | Towards efficient and objective work sampling: Recognizing workers’ activities in site surveillance videos with two-stream convolutional networks [5] |
6 | Dairy goat detection based on Faster R-CNN from surveillance video [6] |
7 | Performance evaluation of deep feature learning for RGB-D image/video classification [7] |
8 | Surveillance scene representation and trajectory abnormality detection using aggregation of multiple concepts [8] |
9 | Human Action Recognition using 3D convolutional neural networks with 3D Motion Cuboids in Surveillance Videos [9] |
10 | Neural networks based visual attention model for surveillance videos [10] |
11 | Application of deep learning for object detection [11] |
12 | A study of deep convolutional auto-encoders for anomaly detection in videos [12] |
13 | A novel deep multi-channel residual networks-based metric learning method for moving human localization in video surveillance [13] |
14 | Video surveillance systems-current status and future trends [14] |
15 | Enhancing transportation systems via deep learning: a survey [15] |
16 | Pedestrian tracking by learning deep features [16] |
17 | Action recognition using spatial-optical data organization and sequential learning framework [17] |
18 | Video pornography detection through deep learning techniques and motion information [18] |
19 | Deep learning to frame objects for visual target tracking [19] |
20 | Boosting deep attribute learning via support vector regression for fast moving crowd counting [20] |
21 | D-STC: deep learning with spatio-temporal constraints for train drivers detection from videos [21] |
22 | A robust human activity recognition system using smartphone sensors and deep learning [22] |
23 | Regional deep learning model for visual tracking [23] |
24 | Fog computing enabled cost-effective distributed summarization of surveillance videos for smart cities [24] |
25 | SIFT and tensor based object detection and classification in videos using deep neural networks [25] |
No: of papers | Journal |
---|---|
19 | Neurocomputing |
14 | Pattern Recognition Letters |
11 | Pattern Recognition |
10 | Journal of Visual Communication and Image Representation |
7 | Expert Systems with Applications |
5 | Procedia Computer Science |
Frequency | Keywords |
---|---|
41 | Deep learning |
11 | Video surveillance |
10 | Convolutional neural network |
9 | Action recognition |
7 | Computer vision |
7 | Person re-identification |
6 | Convolutional neural networks |
5 | CNN |
4 | Activity recognition |
4 | Faster R-CNN |
4 | Machine learning |
4 | Surveillance |
4 | Video |
ACM
Author | Title | Keywords | Journal | Year |
---|---|---|---|---|
Zeng Yu and Tianrui Li and Ning Yu and Yi Pan and Hongmei Chen and Bing Liu | Reconstruction of hidden representation for robust feature extraction [26] | Deep architectures, auto-encoders, feature representation, reconstruction of hidden representation, unsupervised learning | ACM Trans. Intell. Syst. Technol. | 2019 |
Rahim Mammadli and Felix Wolf and Ali Jannesari | The art of getting deep neural networks in shape [27] | Deep neural networks, computer vision, parallel processing | ACM Trans. Archit. Code Optim. | 2019 |
Tinghui Zhou and Richard Tucker and John Flynn and Graham Fyffe and Noah Snavely | Stereo magnification: learning view synthesis using multiplane images [28] | Deep learning, view extrapolation | ACM Trans. Graph. | 2018 |
Zipei Fan and Xuan Song and Tianqi Xia and Renhe Jiang and Ryosuke Shibasaki and Ritsu Sakuramachi | Online deep ensemble learning for predicting citywide human mobility [29] | Deep learning, ensemble learning, human mobility modeling, intelligent surveillance, urban computing | Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. | 2018 |
Rana Hanocka and Noa Fish and Zhenhua Wang and Raja Giryes and Shachar Fleishman and Daniel Cohen-Or | ALIGNet: partial-shape agnostic alignment via unsupervised learning [30] | Deep learning, self-supervised learning, shape deformation | ACM Trans. Graph. | 2018 |
Mengwei Xu and Feng Qian and Qiaozhu Mei and Kang Huang and Xuanzhe Liu | DeepType: on-device deep learning for input personalization service with minimal privacy concern [31] | Deep Learning, Mobile Computing, Personalization | Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. | 2018 |
Thomas E. Potok and Catherine Schuman and Steven Young and Robert Patton and Federico Spedalieri and Jeremy Liu and Ke-Thia Yao and Garrett Rose and Gangotree Chakma | A study of complex deep learning networks on high-performance, neuromorphic, and quantum computers [32] | Deep learning, high-performance computing, neuromorphic computing, quantum computing | J. Emerg. Technol. Comput. Syst. | 2018 |
Samira Pouyanfar and Saad Sadiq and Yilin Yan and Haiman Tian and Yudong Tao and Maria Presa Reyes and Mei-Ling Shyu and Shu-Ching Chen and S. S. Iyengar | A survey on deep learning: algorithms, techniques, and applications [33] | Deep learning, big data, distributed processing, machine learning, neural networks, survey | ACM Comput. Surv. | 2018 |
Yonglong Tian and Guang-He Lee and Hao He and Chen-Yu Hsu and Dina Katabi | RF-based fall monitoring using convolutional neural networks [34] | Deep learning, Device–free, Fall Detection | Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. | 2018 |
Probir Roy and Shuaiwen Leon Song and Sriram Krishnamoorthy and Abhinav Vishnu and Dipanjan Sengupta and Xu Liu | NUMA-Caffe: NUMA-aware deep learning neural networks [35] | Deep learning, NUMA, neural network, stochastic gradient descent | ACM Trans. Archit. Code Optim. | 2018 |
Charles Lovering and Anqi Lu and Cuong Nguyen and Huyen Nguyen and David Hurley and Emmanuel Agu | Fact or fiction [36] | Deep learning, natural language processing, sentiment analysis, social collaboration, subjectivity classification, text classification, web system | Proc. ACM Hum.-Comput. Interact. | 2018 |
Heli Ben-Hamu and Haggai Maron and Itay Kezurer and Gal Avineri and Yaron Lipman | Multi-chart generative surface modeling [37] | Deep learning, generative adveserial networks, shape generation | ACM Trans. Graph. | 2018 |
Weifeng Ge and Bingchen Gong and Yizhou Yu | Image super-resolution via deterministic-stochastic synthesis and local statistical rectification [38] | Deep learning, deterministic component, image superresolution, local correlation matrix, local gram matrix, stochastic component | ACM Trans. Graph. | 2018 |
Peter Hedman and Julien Philip and True Price and Jan-Michael Frahm and George Drettakis and Gabriel Brostow | Deep blending for free-viewpoint image-based rendering [39] | Deep learning, free-viewpoint, image-based rendering | ACM Trans. Graph. | 2018 |
Kalaivani Sundararajan and Damon L. Woodard | Deep learning for biometrics: a survey [40] | Deep learning, autoencoders, convolutional neural networks, deep belief nets, face recognition, feature learning, speaker recognition | ACM Comput. Surv. | 2018 |
Hyungjun Kim and Taesu Kim and Jinseok Kim and Jae-Joon Kim | Deep neural network optimized to resistive memory with nonlinear current–voltage characteristics [41] | Deep neural network, I-V nonlinearity, nonvolatile memory, perceptron | J. Emerg. Technol. Comput. Syst. | 2018 |
Cheng Wang and Haojin Yang and Christoph Meinel | Image captioning with deep bidirectional LSTMs and multi-task learning [42] | Deep learning, LSTM, image captioning, multimodal representations, mutli-task learning | ACM Trans. Multimedia Comput. Commun. Appl. | 2018 |
Shuochao Yao and Yiran Zhao and Huajie Shao and Aston Zhang and Chao Zhang and Shen Li and Tarek Abdelzaher | RDeepSense: reliable deep mobile computing models with uncertainty estimations [43] | Deep Learning, Internet-of-Things, Mobile Computing, Reliability, Uncertainty Estimation | Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. | 2018 |
Dongyu Liu and Weiwei Cui and Kai Jin and Yuxiao Guo and Huamin Qu | DeepTracker: visualizing the training process of convolutional neural networks [44] | Deep learning, correlation analysis, multiple time series, training process, visual analytics | ACM Trans. Intell. Syst. Technol. | 2018 |
Li Yi and Haibin Huang and Difan Liu and Evangelos Kalogerakis and Hao Su and Leonidas Guibas | Deep part induction from articulated object pairs [45] | Deep learning, differentiable sequential RANSAC, motion based part segmentation, shape correspondences | ACM Trans. Graph. | 2018 |
Nanxuan Zhao and Ying Cao and Rynson W. H. Lau | What characterizes personalities of graphic designs? [46] | Deep learning, graphic design, personality | ACM Trans. Graph. | 2018 |
Jiwei Tan and Xiaojun Wan and Hui Liu and Jianguo Xiao | QuoteRec: toward quote recommendation for writing [47] | Deep learning, LSTM, document recommendation, quote recommendation | ACM Trans. Inf. Syst. | 2018 |
Yanru Qu and Bohui Fang and Weinan Zhang and Ruiming Tang and Minzhe Niu and Huifeng Guo and Yong Yu and Xiuqiang He | Product-based neural networks for user response prediction over multi-field categorical data [48] | Deep learning, product-based neural network, recommender system | ACM Trans. Inf. Syst. | 2018 |
Kangxue Yin and Hui Huang and Daniel Cohen-Or and Hao Zhang | P2P-NET: bidirectional point displacement net for shape transform [49] | Deep neural network, point cloud processing, point set transform, point-wise displacement | ACM Trans. Graph. | 2018 |
Shuochao Yao and Yiran Zhao and Huajie Shao and Chao Zhang and Aston Zhang and Shaohan Hu and Dongxin Liu and Shengzhong Liu and Lu Su and Tarek Abdelzaher | SenseGAN: enabling deep learning for internet of things with a semi-supervised framework [50] | Deep Learning, GAN, Internet-of-Things, Mobile Computing, Semi-Supervised Learning | Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. | 2018 |
Shunsuke Saito and Liwen Hu and Chongyang Ma and Hikaru Ibayashi and Linjie Luo and Hao Li | 3D hair synthesis using volumetric variational autoencoders [51] | Deep generative model, hair synthesis, single-view modeling, volumetric variational autoencoder | ACM Trans. Graph. | 2018 |
Anpei Chen and Minye Wu and Yingliang Zhang and Nianyi Li and Jie Lu and Shenghua Gao and Jingyi Yu | Deep surface light fields [52] | Deep Neural Network, Image-based Rendering, Real-time Rendering | Proc. ACM Comput. Graph. Interact. Tech. | 2018 |
IEEE Xplore
Document title | Publication_Year | Funding information |
---|---|---|
Sparse coding guided spatiotemporal feature learning for abnormal event detection in large videos [53] | 2019 | National Nature Science Foundation of China; National Youth Top-notch Talent Support Program |
Rejecting motion outliers for efficient crowd anomaly detection [54] | 2019 | Ministry of Science, ICT and Future Planning |
Deep multi-view feature learning for person re-identification [55] | 2018 | National Natural Science Foundation of China; Yunnan Natural Science Funds; Guangdong Natural Science Funds; Yunnan University |
Image-to-video person re-identification with temporally memorized similarity learning [56] | 2018 | National Natural Science Foundation of China; NSFC-Shenzhen Robotics Projects; Natural Science Foundation of Guangdong Province; Fundamental Research Funds for the Central Universities; ZTE Corporation |
Fight recognition in video using hough forests and 2D convolutional neural network [57] | 2018 | Ministerio de EconomÃa y Competitividad |
Anomalous sound detection using deep audio representation and a BLSTM network for audio surveillance of roads [58] | 2018 | National Natural Science Foundation of China; National Laboratory of Pattern Recognition |
Convolutional neural networks based fire detection in surveillance videos [59] | 2018 | National Research Foundation of Korea (NRF); Korea government (MSIP) |
Action recognition in video sequences using deep bi-directional LSTM with CNN features [60] | 2018 | National Research Foundation of Korea Grant; Korea Government (MSIP) |
A deep spatiotemporal perspective for understanding crowd behavior [61] | 2018 | |
Road traffic conditions classification based on multilevel filtering of image content using convolutional neural networks [62] | 2018 | |
Indoor person identification using a low-power FMCW radar [63] | 2018 | Ghent University; imec; Fund for Scientific Research-Flanders (FWO-Flanders) |
Support vector machine approach to fall recognition based on simplified expression of human skeleton action and fast detection of start key frame using torso angle [64] | 2018 | |
Person re-identification using hybrid representation reinforced by metric learning [65] | 2018 | |
Evolving head tracking routines with brain programming [66] | 2018 | Consejo Nacional de Ciencia y Tecnología; https://doi.org/10.13039/501100004963-seventh Framework Programme of the European Union through the Marie Curie International Research Staff Scheme, FP-PEOPLE-2013-IRSES, Project Analysis and Classification of Mental States of Vigilance with Evolutionary Computation; https://doi.org/10.13039/501100003089-centro de Investigación Científica y de Educación Superior de Ensenada, Baja California; TecNM Project 6474.18-P, “Navegación de robots móviles como un sistema adaptativo complejo.” |
Natural language description of video streams using task-specific feature encoding [67] | 2018 | Basic Science Research Program through the National Research Foundation of Korea (NRF); Ministry of Education |
Background subtraction using multiscale fully convolutional network [68] | 2018 | National Science Foundation of China |
Face verification via learned representation on feature-rich video frames [69] | 2017 | MEITY, India, NVIDIA GPU grant, and Infosys CAI, IIIT-Delhi; IBM Ph.D. fellowship |
Violent activity detection with transfer learning method [70] | 2017 | |
Unsupervised sequential outlier detection with deep architectures [71] | 2017 | |
High-level feature extraction for classification and person re-identification [72] | 2017 | |
An ensemble of invariant features for person reidentification [73] | 2017 | |
Facial expression recognition using salient features and convolutional neural network [74] | 2017 | Research Council of Norway as a part of the Multimodal Elderly Care Systems Project |
Deep head pose: gaze-direction estimation in multimodal video [75] | 2015 | |
Deep reconstruction models for image set classification [76] | 2015 | SIRF; University of Western Australia; ARC |
Violence detection among crowd
Title | Year | Digital repository |
---|---|---|
A review on classifying abnormal behavior in crowd scene [77] | 2019 | ScienceDirect |
Crowd behavior analysis from fixed and moving cameras [78] | 2019 | |
Zero-shot crowd behavior recognition [79] | 2017 | |
The analysis of high density crowds in videos [80] | 2017 | |
Computer vision based crowd disaster avoidance system: a survey [81] | 2017 | |
Deep learning for scene-independent crowd analysis [82] | 2017 | |
Fast face detection in violent video scenes [83] | 2016 | |
Rejecting motion outliers for efficient crowd anomaly detection [54] | 2019 | IEEEXplore |
Deep metric learning for crowdedness regression [84] | 2018 | |
A deep spatiotemporal perspective for understanding crowd behavior [61] | 2018 | |
Crowded scene understanding by deeply learned volumetric slices [85] | 2017 | |
Crowd scene understanding from video: a survey [86] | 2017 | ACM |
Introduction
Surveillance video analysis: relevance in present world
Application areas identified
Surveillance video data as Big Data
Dataset | Type/purpose | Model/schema used |
---|---|---|
ImageNet2012 | Images | |
PASCAL VOC | Images | |
Frames Labeled In Cinema (FLIC) | Popular holywood movies | |
Leeds Sports Pose (LSP) | Sports people gathered from FLICKR | |
CAVIAR | Used for event detection of surveillance domain | Threshold Model used for spatio temporal motion analysis and Bag of Actions for reducing search space [1] |
BEHAVE | Used for event detection of surveillance domain | Threshold Model used for spatio temporal motion analysis and Bag of Actions for reducing search space [1] |
YTO | Videos collected from YouTube | |
i-LIDS sterile zone | People detection | Intrusion detection system with global features [91] |
PETS 2001 | Images | Intrusion detection system with global features [91] |
MoSIFT | Movie dataset | |
STIP | Hockey dataset | |
MediaEval 2013 dataset | Collection of movies | |
UCSD pedestrian | Pedestrian walkway | Convolutional auto-encoder model [12] |
Methods identified/reviewed other than deep learning
-
Event model learning
-
Action model learning
-
Action detection
-
Complex event model learning
-
Complex event detection
-
Pre processing
-
Feature extraction
-
Object tracking
-
Behavior understanding
-
Object and group tracking
-
Grid based analysis
-
Trajectory filtering
-
Abnormal behavior detection using actions descriptors
Title | Method | Tool | Data set |
---|---|---|---|
Scenario-based query processing for video-surveillance archives [95] | Query processing system and inverted tracking | VSQL | PETS 2006 and PETS 2007 |
Activity retrieval in large surveillance videos [96] | Dynamic matching algorithm | Query creation GUI | Pets, Mit traffic |
Integrated video object tracking with applications in trajectory-based event detection [97] | Adaptive particle sampling and Kalman filtering | Not mentioned | PETS 2001 test dataset1, camera 1 |
Evidential event inference in transport video surveillance [98] | Using spatio-temporal correlations for reasoning | Jones and Viola face detector | Own data set |
Abnormal event detection based on analysis of movement information of video sequence [99] | Optical flow and Hidden Markov model | Not mentioned | UMN, PETS |
Anomalous entities detection and localization in pedestrian flows [100] | Gaussian kernel based feature integration and R-CRF model based classification | Not mentioned | UCSD, UMN, UCD |
Snatch theft detection in unconstrained surveillance videos using action attribute modelling [101] | A large GMM called universal attribute model | Own Dataset Snatch 1.0 | |
ArchCam: real time expert system for suspicious behaviour detection in ATM site [102] | Image processing technique | NVIDIA Tegra TX1 SoC 340 with quad core ARM processor and 256 cores GPU | Videos under a mock ATM setup |
-
Object detection
-
Object discrimination
-
Action recognition
Real-time processing in video analysis
Deep learning models in surveillance
-
Human subject detection and discrimination
-
A posture classification module
-
An abnormal behavior detection module
-
You only look once (YOLO) network
-
VGG-16 Net
-
Long short-term memory (LSTM)
Title | Deep learning model | Algorithms used |
---|---|---|
A deep convolutional neural network for video sequence background subtraction [110] | CNN | SuBSENSE algorithm, Flux Tensor algorithm |
Tracking people in RGBD videos using deep learning and motion clues [111] | Deep convolutional neural network | Probabilistic tracking algorithm. |
Deep CNN based binary hash video representations for face retrieval [112] | Deep CNN | Low-rank discriminative binary hashing, back-propagation (BP) algorithm |
DAAL: deep activation-based attribute learning for action recognition in depth videos [113] | 1D temporal CNN, 2D spatial CNN, 3D volumetric CNN | Deep activation-based attribute learning algorithm (DAAL) |
Review in the field of crowd analysis
-
Large number of pedestrians
-
Close proximity
-
Volatility of individual appearance
-
Frequent partial occlusions
-
Irregular motion pattern in crowd
-
Dangerous activities like crowd panic
-
Frame level and pixel level detection
-
Feature detection and temporal filtering
-
Image segmentation and blob extraction
-
Activity detection
-
Activity map
-
Activity analysis
-
Alarm
-
Crowd segmentation and detection
-
Crowd tracking
-
Crowd counting
-
Pedestrian travelling time estimation
-
Crowd attribute recognition
-
Crowd behavior analysis
-
Abnormality detection in a crowd
-
GMM
-
GMM and Markov random field
-
Gaussian poisson mixture model and
-
GMM and support vector machine
-
GM-HMM
-
SLT-HMM
-
MOHMM
-
HM and OSVMs
Title | Method | Tool | Data set |
---|---|---|---|
Measurement of congestion and intrinsic risk in pedestrian crowds [116] | Use computational mesh | Not mentioned | Not mentioned |
A classification method based on streak flow for abnormal crowd behaviors [117] | Streak flow based on fluid mechanics, ViBe algorithm, classification method, | Streakline | ViF |
An intelligent decision computing paradigm for crowd monitoring in the smart city [118] | Extended Kalman filtering approach, Agent motion-based learning model, SIFT feature descriptor, EM algorithm | Not mentioned | The dataset is prepared with surveillance cameras using 60 mm × 120 mm lens from Puri rath yatra festival |
Learning deep event models for crowd anomaly detection [119] | Deep neural network, PCANet, deep GMM | Not mentioned | UCSD Ped1 Dataset, Avenue Dataset |
Results observed from the survey and future directions
-
Time complexity
-
Bad weather conditions
-
Real world dynamics
-
Occulsions
-
Overlapping of objects