Learning cast shadow appearance for human posture recognition

Gouiaa, Rafik; Meunier, Jean

doi:10.1016/j.patrec.2017.06.023

Cited by 4 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a breakthrough in artificial intelligence (AI), deep learning (DL) has overcome previous limitations. DL methods have demonstrated outstanding performances in many fields, such as agriculture (Gouiaa & Meunier 2017;Yang et al 2018), natural language processing (Li 2018), medicine (Gulshan et al 2016), meteorology (Mao et al 2019), bioinformatics (Min et al 2017) and security monitoring (Dhiman & Vishwakarma 2019). DL belongs to the field of machine learning but improves data processing by extracting highly nonlinear and complex features via sequences of multiple layers automatically rather than requiring handcrafted optimal feature representations for a particular type of data based on domain knowledge (LeCun et al 2015;Goodfellow et al 2016).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning for smart fish farming: applications, opportunities and challenges

Yang

Zhang

Liu

et al. 2020

Reviews in Aquaculture

210

View full text Add to dashboard Cite

The rapid emergence of deep learning (DL) technology has resulted in its successful use in various fields, including aquaculture. DL creates both new opportunities and a series of challenges for information and data processing in smart fish farming. This paper focuses on applications of DL in aquaculture, including live fish identification, species classification, behavioural analysis, feeding decisions, size or biomass estimation, and water quality prediction. The technical details of DL methods applied to smart fish farming are also analysed, including data, algorithms and performance. The review results show that the most significant contribution of DL is its ability to automatically extract features. However, challenges still exist; DL is still in a weak artificial intelligence stage and requires large amounts of labelled data for training, which has become a bottleneck that restricts further DL applications in aquaculture. Nevertheless, DL still offers breakthroughs for addressing complex data in aquaculture. In brief, our purpose is to provide researchers and practitioners with a better understanding of the current state of the art of DL in aquaculture, which can provide strong support for implementing smart fish farming applications.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deep learning for smart fish farming: applications, opportunities and challenges

Yang

Zhang

Liu

et al. 2020

Reviews in Aquaculture

210

View full text Add to dashboard Cite

show abstract

“…Activity identification requires translating low-level sensor data to higher-level abstractions (Sun and Wang, 2018). Vector machine-based classification, neural network-based classification, and pattern mating-based classification are the most common algorithms (Neili, et al, 2017;Gouiaa and Meunier, 2017;Hassan, et al, 2018).…”

Section: B Acceleration-based Methodsmentioning

confidence: 99%

Human Body Posture Recognition Approaches

Ali

Hussain

Sadiq

2022

ARO

View full text Add to dashboard Cite

Human body posture recognition has become the focus of many researchers in recent years. Recognition of body posture is used in various applications, including surveillance, security, and health monitoring. However, these systems that determine the body’s posture through video clips, images, or data from sensors have many challenges when used in the real world. This paper provides an important review of how most essential ‎ hardware technologies are ‎used in posture recognition systems‎. These systems capture and collect datasets through ‎accelerometer sensors or computer vision. In addition, this paper presents a comparison ‎study with state-of-the-art in terms of accuracy. We also present the advantages and ‎limitations of each system and suggest promising future ideas that can increase the ‎efficiency of the existing posture recognition system. Finally, the most common datasets ‎applied in these systems are described in detail. It aims to be a resource to help choose one of the methods in recognizing the posture of the human body and the techniques that suit each method. It analyzes more than 80 papers between 2015 and 2020

show abstract

“…However, although this method performs predictive analysis on the behavior of people in the video, using deep convolutional networks, the trade-offs in computational consumption and real-time performance are not fully considered, meanwhile showing that human pose estimation is an important research field of computer vision, and that human key-point detection is a front-end research of human pose estimation. In [4] it was illustrated that human body pose recognition is performed by comparing the shadow of the projection with the shadow of the human body under special circumstances, and proposed a normalization technique to bridge the gap and help the classifier better generalize with real data. Zhang et al [5] proposed three effective training strategies, and exploited four useful postprocessing techniques and proposed a cascaded context mixer (CCM).…”

Section: Introductionmentioning

confidence: 99%

“…Substantial research has been done before in human key-point detection. The purpose of human key-point detection is to estimate the key points of a human body from pictures or videos; it is also an important link in some downstream applications prior to preprocessing, e.g., [4,[7][8][9][10][11]. At present, convolutional neural networks show strong advantages in feature extraction.…”

Section: Introductionmentioning

confidence: 99%

PGNet: Pipeline Guidance for Human Key-Point Detection

Hong

Liu

et al. 2020

Entropy

View full text Add to dashboard Cite

Human key-point detection is a challenging research field in computer vision. Convolutional neural models limit the number of parameters and mine the local structure, and have made great progress in significant target detection and key-point detection. However, the features extracted by shallow layers mainly contain a lack of semantic information, while the features extracted by deep layers contain rich semantic information but a lack of spatial information that results in information imbalance and feature extraction imbalance. With the complexity of the network structure and the increasing amount of computation, the balance between the time of communication and the time of calculation highlights the importance. Based on the improvement of hardware equipment, network operation time is greatly improved by optimizing the network structure and data operation methods. However, as the network structure becomes deeper and deeper, the communication consumption between networks also increases, and network computing capacity is optimized. In addition, communication overhead is also the focus of recent attention. We propose a novel network structure PGNet, which contains three parts: pipeline guidance strategy (PGS); Cross-Distance-IoU Loss (CIoU); and Cascaded Fusion Feature Model (CFFM).

show abstract

Learning cast shadow appearance for human posture recognition

Cited by 4 publications

References 26 publications

Deep learning for smart fish farming: applications, opportunities and challenges

Deep learning for smart fish farming: applications, opportunities and challenges

Human Body Posture Recognition Approaches

PGNet: Pipeline Guidance for Human Key-Point Detection

Contact Info

Product

Resources

About