Video Foreground Extraction Using Multi-View Receptive Field and Encoder–Decoder DCNN for Traffic and Surveillance Applications

Akilan, Thangarajah; Wu, Q. M. Jonathan; Zhang, Wandong

doi:10.1109/tvt.2019.2937076

Cited by 60 publications

(21 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, some studies have proposed the addition of well-designed auxiliary blocks or layers to enhance the motion-related representation capability of the network. The following statistical auxiliary blocks have been used in the literature: SuBSENSE [68], [89], [92], [139], [144], IUTIS [66], [152], PAWCS [83], designed algorithm [91 [141], [145], temporal median [70], [72], [74], [85], [87], [153], [156], [157], temporal histogram and motion saliency map [73], [146], [163], optical flow [127], [155], [164], [165], and conditional random fields (CRF) [67], [69].…”

Section: B Deep Learning Based Methodsmentioning

confidence: 99%

See 1 more Smart Citation

An Empirical Review of Deep Learning Frameworks for Change Detection: Model Design, Experimental Frameworks, Challenges and Research Needs

Mandal¹,

Vipparthi²

2021

Preprint

View full text Add to dashboard Cite

Visual change detection, aiming at segmentation of video frames into foreground and background regions, is one of the elementary tasks in computer vision and video analytics. The applications of change detection include anomaly detection, object tracking, traffic monitoring, human machine interaction, behavior analysis, action recognition, and visual surveillance. Some of the challenges in change detection include background fluctuations, illumination variation, weather changes, intermittent object motion, shadow, fast/slow object motion, camera motion, heterogeneous object shapes and realtime processing. Traditionally, this problem has been solved using hand-crafted features and background modelling techniques. In recent years, deep learning frameworks have been successfully adopted for robust change detection. This article aims to provide an empirical review of the state-of-the-art deep learning methods for change detection. More specifically, we present a detailed analysis of the technical characteristics of different model designs and experimental frameworks. We provide model design based categorization of the existing approaches, including the 2D-CNN, 3D-CNN, ConvLSTM, multi-scale features, residual connections, autoencoders and GAN based methods. Moreover, an empirical analysis of the evaluation settings adopted by the existing deep learning methods is presented. To the best of our knowledge, this is a first attempt to comparatively analyze the different evaluation frameworks used in the existing deep change detection methods. Finally, we point out the research needs, future directions and draw our own conclusions.

show abstract

Section: B Deep Learning Based Methodsmentioning

confidence: 99%

“…Lim et al [91] and Tao et al [141] have also augmented designed blocks to encode the historical patterns. The temporal median has been quite frequently used as a simple temporal feature encoder in [70], [72], [74], [85], [87], [153], [156], [157]. Fig.…”

Section: B Deep Learning Based Methodsmentioning

confidence: 99%

An Empirical Review of Deep Learning Frameworks for Change Detection: Model Design, Experimental Frameworks, Challenges and Research Needs

Mandal¹,

Vipparthi²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The weights of deconvolutional layers keep updating and refining during network training. It is achieved by adding zeros between the consecutive neurons in the receptive field at the input side, and then one convolution kernel is utilized on the top with unit stride (Akilan et al, 2019).…”

Section: Deconvolutional Layersmentioning

confidence: 99%

Deconvolution-and-convolution Networks

Yang,

Zhang,

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

2D Convolutional neural network (CNN) has arguably become the de facto standard for computer vision tasks. Recent findings, however, suggest that CNN may not be the best option for 1D pattern recognition, especially for datasets with over 1 M training samples, e.g., existing CNN-based methods for 1D signals are highly reliant on human pre-processing. Common practices include utilizing discrete Fourier transform (DFT) to reconstruct 1D signal into 2D array. To add to extant knowledge, in this paper, a novel 1D data processing algorithm is proposed for 1D big data analysis through learning a deep deconvolutional-convolutional network. Rather than resorting to human-based techniques, we employed deconvolution layers to convert 1 D signals into 2D data. On top of the deconvolution model, the data was identified by a 2D CNN. Compared with the existing 1D signal processing algorithms, DCNet boasts the advantages of less human-made inference and higher generalization performance. Our experimental results from a varying number of training patterns (50 K to 11 M ) from classification and regression demonstrate the desirability of our new approach.

show abstract

“…Oh et al [ 17 ] proposed multiscale convolutional recurrent neural network for inspecting and classifying bearing fault defects. The literatures [ 18 , 19 ] used multiview receptive field network for foreground detection. Owing to the camera position and angle, objects multiscale features extraction can effectively improve the performance of target detection.…”

Section: Introductionmentioning

confidence: 99%

Integrated Multiscale Appearance Features and Motion Information Prediction Network for Anomaly Detection

Liu

Zhang

Wei

2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The rise of video-prediction algorithms has largely promoted the development of anomaly detection in video surveillance for smart cities and public security. However, most current methods relied on single-scale information to extract appearance (spatial) features and lacked motion (temporal) continuity between video frames. This can cause a loss of partial spatiotemporal information that has great potential to predict future frames, affecting the accuracy of abnormality detection. Thus, we propose a novel prediction network to improve the performance of anomaly detection. Due to the objects of various scales in each video, we use different receptive fields to extract detailed appearance features by the hybrid dilated convolution (HDC) module. Meanwhile, the deeper bidirectional convolutional long short-term memory (DB-ConvLSTM) module can remember the motion information between consecutive frames. Furthermore, we use RGB difference loss to replace optical flow loss as temporal constraint, which greatly reduces the time for optical flow extraction. Compared with the state-of-the-art methods in the anomaly-detection task, experiments prove that our method can more accurately detect abnormalities in various video surveillance scenes.

show abstract

Video Foreground Extraction Using Multi-View Receptive Field and Encoder–Decoder DCNN for Traffic and Surveillance Applications

Cited by 60 publications

References 73 publications

An Empirical Review of Deep Learning Frameworks for Change Detection: Model Design, Experimental Frameworks, Challenges and Research Needs

An Empirical Review of Deep Learning Frameworks for Change Detection: Model Design, Experimental Frameworks, Challenges and Research Needs

Deconvolution-and-convolution Networks

Integrated Multiscale Appearance Features and Motion Information Prediction Network for Anomaly Detection

Contact Info

Product

Resources

About