scholarly journals Towards Fast and Accurate Multi-Person Pose Estimation on Mobile Devices

2021 ◽  
Author(s):  
Xuan Shen ◽  
Geng Yuan ◽  
Wei Niu ◽  
Xiaolong Ma ◽  
Jiexiong Guan ◽  
...  
Keyword(s):  
Mobile Devices ◽  
Pose Estimation ◽  
Rapid Development ◽  
Autonomous Driving ◽  
Abnormal Behavior ◽  
Mobile Platforms ◽  
Abnormal Behavior Detection ◽  
Increasing Demand ◽  
Weight Pruning ◽  
And Behavior

The rapid development of autonomous driving, abnormal behavior detection, and behavior recognition makes an increasing demand for multi-person pose estimation-based applications, especially on mobile platforms. However, to achieve high accuracy, state-of-the-art methods tend to have a large model size and complex post-processing algorithm, which costs intense computation and long end-to-end latency. To solve this problem, we propose an architecture optimization and weight pruning framework to accelerate inference of multi-person pose estimation on mobile devices. With our optimization framework, we achieve up to 2.51X faster model inference speed with higher accuracy compared to representative lightweight multi-person pose estimator.

scholarly journals Abnormal Behavior Detection in Uncrowded Videos with Two-Stream 3D Convolutional Neural Networks

2021 ◽  
Vol 11 (8) ◽  
pp. 3523
Author(s):  
Abid Mehmood
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Video Data ◽  
Abnormal Behavior ◽  
Surveillance Systems ◽  
Detection Rates ◽  
Abnormal Behavior Detection ◽  
General Action ◽  
Increasing Demand ◽  
Abnormal Events

The increasing demand for surveillance systems has resulted in an unprecedented rise in the volume of video data being generated daily. The volume and frequency of the generation of video streams make it both impractical as well as inefficient to manually monitor them to keep track of abnormal events as they occur infrequently. To alleviate these difficulties through intelligent surveillance systems, several vision-based methods have appeared in the literature to detect abnormal events or behaviors. In this area, convolutional neural networks (CNNs) have also been frequently applied due to their prevalence in the related domain of general action recognition and classification. Although the existing approaches have achieved high detection rates for specific abnormal behaviors, more inclusive methods are expected. This paper presents a CNN-based approach that efficiently detects and classifies if a video involves the abnormal human behaviors of falling, loitering, and violence within uncrowded scenes. The approach implements a two-stream architecture using two separate 3D CNNs to accept a video and an optical flow stream as input to enhance the prediction performance. After applying transfer learning, the model was trained on a specialized dataset corresponding to each abnormal behavior. The experiments have shown that the proposed approach can detect falling, loitering, and violence with an accuracy of up to 99%, 97%, and 98%, respectively. The model achieved state-of-the-art results and outperformed the existing approaches.

The Oxford Handbook of Ethics of AI

2020 ◽  
Keyword(s):  
Artificial Intelligence ◽  
Ethical Issues ◽  
Rapid Development ◽  
Data Driven ◽  
International Human Rights Law ◽  
Human Knowledge ◽  
Human Rights Law ◽  
Ai Ethics ◽  
And Behavior ◽  
Control Complex

This book explores the intertwining domains of artificial intelligence (AI) and ethics—two highly divergent fields which at first seem to have nothing to do with one another. AI is a collection of computational methods for studying human knowledge, learning, and behavior, including by building agents able to know, learn, and behave. Ethics is a body of human knowledge—far from completely understood—that helps agents (humans today, but perhaps eventually robots and other AIs) decide how they and others should behave. Despite these differences, however, the rapid development in AI technology today has led to a growing number of ethical issues in a multitude of fields, ranging from disciplines as far-reaching as international human rights law to issues as intimate as personal identity and sexuality. In fact, the number and variety of topics in this volume illustrate the width, diversity of content, and at times exasperating vagueness of the boundaries of “AI Ethics” as a domain of inquiry. Within this discourse, the book points to the capacity of sociotechnical systems that utilize data-driven algorithms to classify, to make decisions, and to control complex systems. Given the wide-reaching and often intimate impact these AI systems have on daily human lives, this volume attempts to address the increasingly complicated relations between humanity and artificial intelligence. It considers not only how humanity must conduct themselves toward AI but also how AI must behave toward humanity.

scholarly journals Standardized Method for the Assessment of Behavioral Responses of Zebrafish Larvae

Biomedicines ◽  
2021 ◽  
Vol 9 (8) ◽  
pp. 884
Author(s):  
Hideyuki Maeda ◽  
Noritoshi Fukushima ◽  
Akihiro Hasumi
Keyword(s):  
Early Stage ◽  
Scientific Evidence ◽  
Video Tracking ◽  
Abnormal Behavior ◽  
Laboratory Setting ◽  
Zebrafish Larvae ◽  
Experimental Systems ◽  
Standardized Method ◽  
Life Phenomena ◽  
And Behavior

Zebrafish are easy to breed in a laboratory setting as they are extremely fertile and produce dozens of eggs per set. Because zebrafish eggs and the skin of the early-stage larvae are transparent, their embryos and the hearts and muscles of their larvae can be easily observed. Multiple rapid analyses of heart rate and behavior can be performed on these larvae simultaneously, enabling investigation of the influence of neuroactive substances on abnormal behavior, death, and associated pathogenetic mechanisms. Zebrafish larvae are becoming increasingly popular among researchers and are used in laboratories worldwide to study various vertebrate life phenomena; more experimental systems using zebrafish will undoubtedly be developed in the future. However, based on the available literature, we believe that the conceptualization of a protocol based on scientific evidence is necessary to achieve standardization. We exposed zebrafish larvae at 6–7 days post-fertilization to 50 repeated light–dark stimuli at either 15-min or 5-min intervals. We measured the traveled distance and habituation time through a video tracking apparatus. The traveled distance stabilized after the 16th repetition when the zebrafish were exposed to light–dark stimuli at 15-min intervals and after the 5th repetition when exposed at 5-min intervals. Additionally, at 15-min intervals, the peak of the traveled distance was reached within the first minute in a dark environment, whereas at 5-min intervals, it did not reach the peak even after 5 min. The traveled distance was more stable at 5-min intervals of light/dark stimuli than at 15-min intervals. Therefore, if one acclimatizes zebrafish larvae for 1 h and collects data from the 5th repetition of light/dark stimuli at intervals of 5 min in the light/dark test, a stable traveled distance result can be obtained. The establishment of this standardized method would be beneficial for investigating substances of unknown lethal concentration.

scholarly journals End-to-End Autonomous Driving Through Dueling Double Deep Q-Network

2021 ◽  
Author(s):  
Baiyu Peng ◽  
Qi Sun ◽  
Shengbo Eben Li ◽  
Dongsuk Kum ◽  
Yuming Yin ◽  
...  
Keyword(s):  
Neural Network ◽  
State Space ◽  
Learning Algorithm ◽  
Rapid Development ◽  
Autonomous Driving ◽  
Saliency Map ◽  
Hierarchical Architecture ◽  
Link Type ◽  
The Neural Network ◽  
End To End

AbstractRecent years have seen the rapid development of autonomous driving systems, which are typically designed in a hierarchical architecture or an end-to-end architecture. The hierarchical architecture is always complicated and hard to design, while the end-to-end architecture is more promising due to its simple structure. This paper puts forward an end-to-end autonomous driving method through a deep reinforcement learning algorithm Dueling Double Deep Q-Network, making it possible for the vehicle to learn end-to-end driving by itself. This paper firstly proposes an architecture for the end-to-end lane-keeping task. Unlike the traditional image-only state space, the presented state space is composed of both camera images and vehicle motion information. Then corresponding dueling neural network structure is introduced, which reduces the variance and improves sampling efficiency. Thirdly, the proposed method is applied to The Open Racing Car Simulator (TORCS) to demonstrate its great performance, where it surpasses human drivers. Finally, the saliency map of the neural network is visualized, which indicates the trained network drives by observing the lane lines. A video for the presented work is available online, https://youtu.be/76ciJmIHMD8 or https://v.youku.com/v_show/id_XNDM4ODc0MTM4NA==.html.

scholarly journals Examination of Abnormal Behavior Detection Based on Improved YOLOv3

Electronics ◽  
2021 ◽  
Vol 10 (2) ◽  
pp. 197
Author(s):  
Meng-ting Fang ◽  
Zhong-ju Chen ◽  
Krzysztof Przystupa ◽  
Tao Li ◽  
Michal Majka ◽  
...  
Keyword(s):  
Detection Method ◽  
Abnormal Behavior ◽  
Detection Accuracy ◽  
Average Precision ◽  
Research Results ◽  
Behavior Detection ◽  
Abnormal Behavior Detection ◽  
The Mean ◽  
Examination Room ◽  
Detection Speed

Examination is a way to select talents, and a perfect invigilation strategy can improve the fairness of the examination. To realize the automatic detection of abnormal behavior in the examination room, the method based on the improved YOLOv3 (The third version of the You Only Look Once algorithm) algorithm is proposed. The YOLOv3 algorithm is improved by using the K-Means algorithm, GIoUloss, focal loss, and Darknet32. In addition, the frame-alternate dual-thread method is used to optimize the detection process. The research results show that the improved YOLOv3 algorithm can improve both the detection accuracy and detection speed. The frame-alternate dual-thread method can greatly increase the detection speed. The mean Average Precision (mAP) of the improved YOLOv3 algorithm on the test set reached 88.53%, and the detection speed reached 42 Frames Per Second (FPS) in the frame-alternate dual-thread detection method. The research results provide a certain reference for automated invigilation.

Building pose estimation from the perspective of UAVs based on CNNs

2021 ◽  
pp. 2150003
Author(s):  
Zhenni Wu ◽  
Hengxin Chen ◽  
Bin Fang ◽  
Zihao Li ◽  
Xinrun Chen
Keyword(s):  
Neural Network ◽  
Pose Estimation ◽  
Computer Technology ◽  
Architectural Design ◽  
Rapid Development ◽  
Angle Error ◽  
Building Model ◽  
Building Models ◽  
World Coordinate System ◽  
Aerial Vehicle

With the rapid development of computer technology, building pose estimation combined with Augmented Reality (AR) can play a crucial role in the field of urban planning and architectural design. For example, a virtual building model can be placed into a realistic scenario acquired by a Unmanned Aerial Vehicle (UAV) to visually observe whether the building can integrate well with its surroundings, thus optimizing the design of the building. In the work, we contribute a building dataset for pose estimation named BD3D. To obtain accurate building pose, we use a physical camera which can simulate realistic cameras in Unity3D to simulate UAVs perspective and use virtual building models as objects. We propose a novel neural network that combines MultiBin module with PoseNet architecture to estimate the building pose. Sometimes, the building is symmetry and ambiguity causes its different surfaces to have similar features, making it difficult for CNNs to learn the differential features between the different surfaces. We propose a generalized world coordinate system repositioning strategy to deal with it. We evaluate our network with the strategy on BD3D, and the angle error is reduced to [Formula: see text] from [Formula: see text]. Code and dataset have been made available at: https://github.com/JellyFive/Building-pose-estimation-from-the-perspective-of-UAVs-based-on-CNNs .

scholarly journals A Novel Detection Framework for Detecting Abnormal Human Behavior

2020 ◽  
Vol 2020 ◽  
pp. 1-9
Author(s):  
Chengfei Wu ◽  
Zixuan Cheng
Keyword(s):  
Human Behavior ◽  
Public Safety ◽  
Space Time ◽  
Abnormal Behavior ◽  
Detection Methods ◽  
Time Block ◽  
Safety Issues ◽  
Behavior Detection ◽  
Abnormal Behavior Detection ◽  
Abnormal Behaviors

Public safety issues have always been the focus of widespread concern of people from all walks of life. With the development of video detection technology, the detection of abnormal human behavior in videos has become the key to preventing public safety issues. Particularly, in student groups, the detection of abnormal human behavior is very important. Most existing abnormal human behavior detection algorithms are aimed at outdoor activity detection, and the indoor detection effects of these algorithms are not ideal. Students spend most of their time indoors, and modern classrooms are mostly equipped with monitoring equipment. This study focuses on the detection of abnormal behaviors of indoor humans and uses a new abnormal behavior detection framework to realize the detection of abnormal behaviors of indoor personnel. First, a background modeling method based on a Gaussian mixture model is used to segment the background image of each image frame in the video. Second, block processing is performed on the image after segmenting the background to obtain the space-time block of each frame of the image, and this block is used as the basic representation of the detection object. Third, the foreground image features of each space-time block are extracted. Fourth, fuzzy C-means clustering (FCM) is used to detect outliers in the data sample. The contribution of this paper is (1) the use of an abnormal human behavior detection framework that is effective indoors. Compared with the existing abnormal human behavior detection methods, the detection framework in this paper has a little difference in terms of its outdoor detection effects. (2) Compared with other detection methods, the detection framework used in this paper has a better detection effect for abnormal human behavior indoors, and the detection performance is greatly improved. (3) The detection framework used in this paper is easy to implement and has low time complexity. Through the experimental results obtained on public and manually created data sets, it can be demonstrated that the performance of the detection framework used in this paper is similar to those of the compared methods in outdoor detection scenarios. It has a strong advantage in terms of indoor detection. In summary, the proposed detection framework has a good practical application value.

Sign in / Sign up

Export Citation Format

Share Document