Detection of Customer Interested Garments in Surveillance Video using Computer Vision

Studies have shown that the human capability of monitoring multiple surveillance videos is limited. Computer vision techniques have been developed to detect abnormal events to support human video surveillance; however, their results are often unreliable, thus distracting surveillance operators and making them miss important events. This article presents Hotspot as a surveillance video visualization system that can effectively leverage noisy computer vision techniques to support human video surveillance. Hotspot consists of two views: a designated focus view to summarize videos with detected events and a video-bank view surrounding the focus view to display source surveillance videos. The focus view allows an operator to quickly dismiss false alarms and focus on true alarms. The video-bank view allows for extended human video analysis after an important event is detected. Hotspot further provides visual links to assist quick attention switch from the focus view to the video-bank view. Our experiments show that Hotspot can effectively integrate noisy, automatic computer vision detection results and better support human video surveillance tasks than the baseline video surveillance with no or only basic computer vision support.

Download Full-text

Topic-based Video Analysis

ACM Computing Surveys ◽

10.1145/3459089 ◽

2021 ◽

Vol 54 (6) ◽

pp. 1-34

Author(s):

Ratnabali Pal ◽

Arif Ahmed Sekh ◽

Debi Prosad Dogra ◽

Samarjit Kar ◽

Partha Pratim Roy ◽

...

Keyword(s):

Computer Vision ◽

Video Analysis ◽

Visual Surveillance ◽

Video Data ◽

Camera Motion ◽

Topic Modelling ◽

Surveillance Video ◽

Analysis Computer ◽

Manual Processing ◽

Spatio Temporal

Manual processing of a large volume of video data captured through closed-circuit television is challenging due to various reasons. First, manual analysis is highly time-consuming. Moreover, as surveillance videos are recorded in dynamic conditions such as in the presence of camera motion, varying illumination, or occlusion, conventional supervised learning may not work always. Thus, computer vision-based automatic surveillance scene analysis is carried out in unsupervised ways. Topic modelling is one of the emerging fields used in unsupervised information processing. Topic modelling is used in text analysis, computer vision applications, and other areas involving spatio-temporal data. In this article, we discuss the scope, variations, and applications of topic modelling, particularly focusing on surveillance video analysis. We have provided a methodological survey on existing topic models, their features, underlying representations, characterization, and applications in visual surveillance’s perspective. Important research papers related to topic modelling in visual surveillance have been summarized and critically analyzed in this article.

Download Full-text

Deep neural networks for analysis of fisheries surveillance video and automated monitoring of fish discards

ICES Journal of Marine Science ◽

10.1093/icesjms/fsz149 ◽

2019 ◽

Vol 77 (4) ◽

pp. 1340-1353 ◽

Cited By ~ 1

Author(s):

Geoff French ◽

Michal Mackiewicz ◽

Mark Fisher ◽

Helen Holah ◽

Rachel Kilburn ◽

...

Keyword(s):

Neural Network ◽

Computer Vision ◽

Deep Neural Network ◽

Deep Neural Networks ◽

Vision System ◽

Surveillance Video ◽

Computer Vision System ◽

Species Classification ◽

Fish Catch ◽

Computer Vision Systems

Abstract We report on the development of a computer vision system that analyses video from CCTV systems installed on fishing trawlers for the purpose of monitoring and quantifying discarded fish catch. Our system is designed to operate in spite of the challenging computer vision problem posed by conditions on-board fishing trawlers. We describe the approaches developed for isolating and segmenting individual fish and for species classification. We present an analysis of the variability of manual species identification performed by expert human observers and contrast the performance of our species classifier against this benchmark. We also quantify the effect of the domain gap on the performance of modern deep neural network-based computer vision systems.

Download Full-text

Sex Classification via 2D-Skeletonization

Mathematical Problems in Engineering ◽

10.1155/2020/6182654 ◽

2020 ◽

Vol 2020 ◽

pp. 1-12

Author(s):

Miguel Contreras-Murillo ◽

Sergio G. de-los-Cobos-Silva ◽

Pedro Lara-Velázquez ◽

Eric A. Rincón-García ◽

Román A. Mora-Gutiérrez ◽

...

Keyword(s):

Computer Vision ◽

Open Problem ◽

Surveillance Video ◽

Camera Angle ◽

The Creation ◽

2D Images

Sex classification is a challenging open problem in computer vision. It is useful from statistics up to people recognition on surveillance video. So far, the best performance can be achieved by using 3D cameras, but this approach requires the use of some especial hardware. Other 2D approaches achieve good results on normal situations but fail when the person wears loose clothing and carries bags or the camera angle changes as they rely on calculating borders, silhouettes, or the energy of the person in the image. This work aims to provide a novel sex classification methodology based on the creation of a virtual skeleton for each individual from 2D images and video; then, the distances between some points of the skeleton are measured and work as input of a sex classifier. This improves the results since clothing, bags, and the camera angle affect little the skeletonization process.

Download Full-text

A novel multi-stream method for violent interaction detection using deep learning

Measurement and Control ◽

10.1177/0020294020902788 ◽

2020 ◽

Vol 53 (5-6) ◽

pp. 796-806

Author(s):

Hongchang Li ◽

Jing Wang ◽

Jianjun Han ◽

Jinmin Zhang ◽

Yushan Yang ◽

...

Keyword(s):

Computer Vision ◽

Deep Learning ◽

Optical Flow ◽

High Probability ◽

Recognition Performance ◽

Surveillance Video ◽

Interaction Detection ◽

Temporal Features ◽

Violent Action ◽

Learning In Computer Vision

Violent interaction detection is a hot topic in computer vision. However, the recent research works on violent interaction detection mainly focus on the traditional hand-craft features, and does not make full use of the research results of deep learning in computer vision. In this paper, we propose a new robust violent interaction detection framework based on multi-stream deep learning in surveillance scene. The proposed approach enhances the recognition performance of violent action in video by fusing three different streams: attention-based spatial RGB stream, temporal stream, and local spatial stream. The attention-based spatial RGB stream learns the spatial attention regions of persons that have high probability to be action region through soft-attention mechanism. The temporal stream employs optical flow as input to extract temporal features. The local spatial stream learns spatial local features using block images as input. Experimental results demonstrate the effectiveness and reliability of the proposed method on three violent interactive datasets: hockey fights, movies, violent interaction. We also verify the proposed method on our own elevator surveillance video dataset and the performance of the proposed method is satisfied.

Download Full-text