scholarly journals Hi-EADN: Hierarchical Excitation Aggregation and Disentanglement Frameworks for Action Recognition Based on Videos

Symmetry ◽  
2021 ◽  
Vol 13 (4) ◽  
pp. 662
Author(s):  
Zeyuan Hu ◽  
Eung-Joo Lee
Keyword(s):  
Action Recognition ◽  
Short Range ◽  
Semantic Information ◽  
Long Distance ◽  
Spatiotemporal Feature ◽  
Benchmark Datasets ◽  
Feature Information ◽  
Multiple Frame ◽  
High Level ◽  
Information Streams

Most existing video action recognition methods mainly rely on high-level semantic information from convolutional neural networks (CNNs) but ignore the discrepancies of different information streams. However, it does not normally consider both long-distance aggregations and short-range motions. Thus, to solve these problems, we propose hierarchical excitation aggregation and disentanglement networks (Hi-EADNs), which include multiple frame excitation aggregation (MFEA) and a feature squeeze-and-excitation hierarchical disentanglement (SEHD) module. MFEA specifically uses long-short range motion modelling and calculates the feature-level temporal difference. The SEHD module utilizes these differences to optimize the weights of each spatiotemporal feature and excite motion-sensitive channels. Moreover, without introducing additional parameters, this feature information is processed with a series of squeezes and excitations, and multiple temporal aggregations with neighbourhoods can enhance the interaction of different motion frames. Extensive experimental results confirm our proposed Hi-EADN method effectiveness on the UCF101 and HMDB51 benchmark datasets, where the top-5 accuracy is 93.5% and 76.96%.

scholarly journals Challenges and Limitations in Human Action Recognition on Unmanned Aerial Vehicles: A Comprehensive Survey

2021 ◽  
Vol 38 (5) ◽  
pp. 1403-1411
Author(s):  
Nashwan Adnan Othman ◽  
Ilhan Aydin
Keyword(s):  
Action Recognition ◽  
Smart Cities ◽  
Large Angle ◽  
Human Action Recognition ◽  
Human Action ◽  
Human Detection ◽  
Long Distance ◽  
Benchmark Datasets ◽  
Comprehensive Survey ◽  
Aerial Vehicle

An Unmanned Aerial Vehicle (UAV), commonly called a drone, is an aircraft without a human pilot aboard. Making UAVs that can accurately discover individuals on the ground is very important for various applications, such as people searches, and surveillance. UAV integration in smart cities is challenging, however, because of problems and concerns such as privacy, safety, and ethical/legal use. Human action recognition-based UAVs can utilize modern technologies. Thus, it is essential for future development of the aforementioned applications. UAV-based human activity recognition is the procedure of classifying photo sequences with action labels. This paper offers a comprehensive study of UAV-based human action recognition techniques. Furthermore, we conduct empirical research studies to assess several factors that might influence the efficiency of human detection and action recognition techniques in UAVs. Benchmark datasets commonly utilized for UAV-based human action recognition are briefly explained. Our findings reveal that the existing human action recognition innovations can identify human actions on UAVs with some limitations in range, altitudes, long-distance, and a large angle of depression.

scholarly journals Part-Level Graph Convolutional Network for Skeleton-Based Action Recognition

2020 ◽  
Vol 34 (07) ◽  
pp. 11045-11052
Author(s):  
Linjiang Huang ◽  
Yan Huang ◽  
Wanli Ouyang ◽  
Liang Wang
Keyword(s):  
Action Recognition ◽  
State Of The Art ◽  
The State ◽  
Body Parts ◽  
Convolutional Network ◽  
Joint Level ◽  
Convolutional Networks ◽  
Level Information ◽  
Benchmark Datasets ◽  
High Level

Recently, graph convolutional networks have achieved remarkable performance for skeleton-based action recognition. In this work, we identify a problem posed by the GCNs for skeleton-based action recognition, namely part-level action modeling. To address this problem, a novel Part-Level Graph Convolutional Network (PL-GCN) is proposed to capture part-level information of skeletons. Different from previous methods, the partition of body parts is learnable rather than manually defined. We propose two part-level blocks, namely Part Relation block (PR block) and Part Attention block (PA block), which are achieved by two differentiable operations, namely graph pooling operation and graph unpooling operation. The PR block aims at learning high-level relations between body parts while the PA block aims at highlighting the important body parts in the action. Integrating the original GCN with the two blocks, the PL-GCN can learn both part-level and joint-level information of the action. Extensive experiments on two benchmark datasets show the state-of-the-art performance on skeleton-based action recognition and demonstrate the effectiveness of the proposed method.

scholarly journals Exploring 3D Human Action Recognition Using STACOG on Multi-View Depth Motion Maps Sequences

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3642
Author(s):  
Mohammad Farhad Bulbul ◽  
Sadiya Tabussum ◽  
Hazrat Ali ◽  
Wenli Zheng ◽  
Mi Young Lee ◽  
...  
Keyword(s):  
Action Recognition ◽  
Depth Map ◽  
Human Action Recognition ◽  
Human Action ◽  
Collaborative Representation ◽  
Auto Correlation ◽  
Time Operation ◽  
Real Time Operation ◽  
Benchmark Datasets ◽  
Depth Motion Maps

This paper proposes an action recognition framework for depth map sequences using the 3D Space-Time Auto-Correlation of Gradients (STACOG) algorithm. First, each depth map sequence is split into two sets of sub-sequences of two different frame lengths individually. Second, a number of Depth Motion Maps (DMMs) sequences from every set are generated and are fed into STACOG to find an auto-correlation feature vector. For two distinct sets of sub-sequences, two auto-correlation feature vectors are obtained and applied gradually to L2-regularized Collaborative Representation Classifier (L2-CRC) for computing a pair of sets of residual values. Next, the Logarithmic Opinion Pool (LOGP) rule is used to combine the two different outcomes of L2-CRC and to allocate an action label of the depth map sequence. Finally, our proposed framework is evaluated on three benchmark datasets named MSR-action 3D dataset, DHA dataset, and UTD-MHAD dataset. We compare the experimental results of our proposed framework with state-of-the-art approaches to prove the effectiveness of the proposed framework. The computational efficiency of the framework is also analyzed for all the datasets to check whether it is suitable for real-time operation or not.

Are the World Junior Championship Finalists for Middle- and Long-Distance Events Currently Competing at International Level?

2017 ◽  
Vol 12 (3) ◽  
pp. 316-321 ◽  
Author(s):  
Federico Pizzuto ◽  
Matteo Bonato ◽  
Gialunca Vernillo ◽  
Antonio La Torre ◽  
Maria Francesca Piacentini
Keyword(s):  
Olympic Games ◽  
International Association ◽  
Age Effect ◽  
Long Distance ◽  
Relative Age ◽  
Relative Age Effect ◽  
The World ◽  
Elite Level ◽  
And Gender ◽  
High Level

Purpose:To analyze how many finalists of the International Association of Athletics Federations (IAAF) World Junior Championships (WJCs) in the middle- and long-distance track events had dropped out from high-level competitions.Methods:Starting from 2002, the 8 male and the 8 female finalists in the middle- and long-distance events of 6 editions of the WJC were followed until 2015 to evaluate how many missed the IAAF rankings for 2 consecutive years starting from the year after WJC participation. For those still competing at elite level, their careers were monitored.Results:In 2015, 61% of the 2002, 54.8% of the 2004, 48.3% of the 2006, 37.5% of the 2008, 26.2% of the 2010, and 29% of the 2012 WJC finalists were not present in the IAAF rankings. Of the 368 athletes considered, 75 (20.4%) were able to achieve the IAAF top 10 in 2.4 ± 2.2 y. There is evidence of relationships between dropout and gender (P = .040), WJC edition (P = .000), and nationality (P = .010) and between the possibility to achieve the IAAF top 10 and dropout (P = .000), continent (P = .001), relative age effect (P = .000), and quartile of birth (P = .050).Conclusions:Even if 23 of the finalists won a medal at the Olympic Games or at the World Championships, it is still not clear if participation at the WJC is a prerequisite to success at a senior level.

scholarly journals Knowledge Integration Networks for Action Recognition

2020 ◽  
Vol 34 (07) ◽  
pp. 12862-12869
Author(s):  
Shiwen Zhang ◽  
Sheng Guo ◽  
Limin Wang ◽  
Weilin Huang ◽  
Matthew Scott
Keyword(s):  
Action Recognition ◽  
Large Scale ◽  
Knowledge Integration ◽  
Scene Recognition ◽  
Teacher Networks ◽  
Medium Level ◽  
Meaningful Context ◽  
Context Knowledge ◽  
High Level ◽  
Context Features

In this work, we propose Knowledge Integration Networks (referred as KINet) for video action recognition. KINet is capable of aggregating meaningful context features which are of great importance to identifying an action, such as human information and scene context. We design a three-branch architecture consisting of a main branch for action recognition, and two auxiliary branches for human parsing and scene recognition which allow the model to encode the knowledge of human and scene for action recognition. We explore two pre-trained models as teacher networks to distill the knowledge of human and scene for training the auxiliary tasks of KINet. Furthermore, we propose a two-level knowledge encoding mechanism which contains a Cross Branch Integration (CBI) module for encoding the auxiliary knowledge into medium-level convolutional features, and an Action Knowledge Graph (AKG) for effectively fusing high-level context information. This results in an end-to-end trainable framework where the three tasks can be trained collaboratively, allowing the model to compute strong context knowledge efficiently. The proposed KINet achieves the state-of-the-art performance on a large-scale action recognition benchmark Kinetics-400, with a top-1 accuracy of 77.8%. We further demonstrate that our KINet has strong capability by transferring the Kinetics-trained model to UCF-101, where it obtains 97.8% top-1 accuracy.

scholarly journals The ascent and decline of a great balneological resort: PUCIOASA

2020 ◽  
pp. 498-500
Author(s):  
Doroteea TEOIBAS-SERBAN ◽  
Liviu IORDACHE ◽  
Dan BLENDEA
Keyword(s):  
Water Analysis ◽  
18Th Century ◽  
Local Level ◽  
Iron Concentration ◽  
Long Distance ◽  
Number Of Patients ◽  
The Government ◽  
High Level ◽  
In The Beginning ◽  
Rehabilitation Clinic

Introduction. Pucioasa is one of the greater and sadly, almost forgotten balneological resort of Romania. Known since the 18th century, where it was discovered a record of this area on the Austrian map in 1791 with sulphurous streams, the healing waters of the “Pe Pucioasa” hill enters the therapeutic balneological circle when the first water analysis occurred – 1821-1828 by Dr. Trangot von Schobel. In 1878, Ion Ghica, a prominent figure in Romania at that time initiated the exploitation of 14 wells from the 30 known streams of sulphurous and ferruginous waters, thus obtaining enough healing mineral water for a capacity of 60 beds of the balneological establishment. Since 1841, the Resort developed, transformed into a Rehabilitation Clinic (1969) and the number of patients treated there increased from 20 in the beginning to 1000 patient per day. Many of the patients included foreign elite society members who came regularly to benefit from the waters’ healing properties, both external and internal cures, setting Pucioasa on a high level of not only balneology treatment, but also of balneological tourism. Matherials and methods: Studying local industrializations and water analysis, we concluded that there were many factors that contributed to the decline of this great and international renowned balneological resort. The first wrong step in this direction was made when building and extending de gypsum carrier, the peek being between 1970 and 1975, when the industrial forging process made the sulphurous streams to migrate, and the waters from the wells diminished in its sulphur and iron concentration. Another possible factor for the decline can be the modernisation of the medical world, development of anti-inflammatory drugs and modern medical equipment which reduced the need for the elite society to come a long distance for the balneological treatment. Results: Whichever of the reasons, the once prosperous Balneological Resort of Pucioasa fell into an unknown, unimpressive place, known and frequented only by the locals. The Rehabilitation Clinic still has 60 beds, which are occupied to the maximum each month, it still prospers on a local level, but lost its glory from almost a century ago. Conclusions: The Romanian general population and the government must take an interest to one of the many once prosperous balneological resorts in order to revive the national treasure of Romania which is balneology, which many of the developed European countries only wish that they can possess. Keywords: Balneary Resort, Pucioasa,

Hybrid-Mode-Assisted Long-Distance Excitation of Short-Range Surface Plasmons in a Nanotip-Enhanced Step-Index Fiber

Nano Letters ◽  
2017 ◽  
Vol 17 (2) ◽  
pp. 631-637 ◽  
Author(s):  
Alessandro Tuniz ◽  
Mario Chemnitz ◽  
Jan Dellith ◽  
Stefan Weidlich ◽  
Markus A. Schmidt
Keyword(s):  
Surface Plasmons ◽  
Short Range ◽  
Hybrid Mode ◽  
Long Distance ◽  
Step Index ◽  
Step Index Fiber

Obsidian in Southern Ecuador: The Carboncillo Source

2011 ◽  
Vol 22 (1) ◽  
pp. 97-120 ◽  
Author(s):  
Dennis E. Ogburn
Keyword(s):  
Economic Organization ◽  
Geochemical Analysis ◽  
Long Distance ◽  
X Ray ◽  
Far North ◽  
The North ◽  
The Andes ◽  
Southern Ecuador ◽  
High Level ◽  
Northern Highlands

AbstractThe Carboncillo area in the southern highlands of Ecuador is identified as the only confirmed source of archaeological obsidian located in the country outside of the northern highlands and is the first identified in the large gap between the known Ecuadorian and Peruvian sources. With the identification of this source, it can no longer be assumed that all obsidian found in Ecuador came from sources in the northern highlands. Thus the Carboncillo source has significant implications for interpreting patterns of long-distance exchange in the Andes, especially in southern Ecuador and the far north of Peru. A geochemical analysis of the Carboncillo material shows that it can be easily distinguished from the obsidian from other Ecuadorian sources. A provenance study of archaeological obsidian samples from the southern highlands of Ecuador using x-ray fluorescence demonstrates that the Carboncillo obsidian was used at the Preceramic site of Chobshi Cave and at a number of late prehispanic sites in the Saraguro region. The results indicate a high level of sociopolitical and economic isolation in late prehistory, most likely tied to a preoccupation with warfare between neighboring groups, and provide additional evidence that the economic organization of the southern highlands differed markedly from that of the north during this time.

Rural Centres, Trade, and Non-Agrarian Production

2002 ◽  
Author(s):  
Helena Hamerow
Keyword(s):  
State Formation ◽  
Craft Production ◽  
Northern Europe ◽  
Viking Age ◽  
Rural Settlements ◽  
Long Distance ◽  
Farming Communities ◽  
Before And After ◽  
Trade And Exchange ◽  
High Level

In contrast to the relative scarcity of publications dealing with the buildings and layouts of rural settlements, many volumes have been devoted to the development of early medieval trade and craft production (e.g. Jankuhn et al. 1981; 1983; K. Düwel et al. 1987, vols. 1–4; Hodges and Whitehouse 1983). Archaeological research into these topics has been made more fruitful—as well as more complex—by the contributions of neighbouring disciplines such as history, geography, and numismatics. It has, however, tended to focus almost exclusively on towns, monasteries, and royal centres, yet craft production, trade, and exchange also played a significant role in farming communities before and after the emergence of such specialized centres. Indeed, the rural settlements of northwest Europe were already significantly differentiated in their economies in the Migration period, suggesting a high level of socio-economic complexity several centuries earlier than has generally been supposed. The evidence now available for trade and non-agrarian production, which derives almost wholly from archaeology, calls for a thoroughgoing reassessment of when and how centralized authorities emerged in northern Europe after the collapse of the western Empire. This is particularly true for northern Germany and southern Scandinavia, where early state formation has conventionally been dated to the late Viking period. Research into state formation has in the past focused on the origins of towns and market centres, the latter usually seen as arising from participation in long-distance trade which was controlled by kings or magnates. Yet, several centuries before there were kings or towns in northern Europe, rural settlements emerged which point to a degree of political centralization. This chapter considers the evidence for these rural centres and the role of non-agrarian production and exchange in rural settlements generally: what was the scale and context of the production, distribution, and consumption of non-agrarian goods? Who controlled these activities, and how, if at all, did the long-distance trade networks which fuelled the nascent towns of Merovingian and Viking Age Europe affect the economies of the communities which lay in their hinterlands?

A High-Throughput Architecture for the SHA-256/224 Compliant With the DSRC Standard

2019 ◽  
Vol 10 (1) ◽  
pp. 98-118
Author(s):  
Imed Saad Ben Dhaou ◽  
Hannu Tenhunen
Keyword(s):  
High Throughput ◽  
Short Range ◽  
State Of The Art ◽  
High Level Synthesis ◽  
Data Flow Graph ◽  
Transformation Techniques ◽  
Safety Message ◽  
Dedicated Short Range Communication ◽  
Field Programmable ◽  
High Level

This article presents a word serial retimed architecture for the SHA-256/224 algorithm. The architecture is compliant with the dedicated-short range communication for safety message authentications. We elaborate three-operand adder architectures suitable for field programmable gate array implementation. Several transformation techniques at the data-flow-graph level have been used to derive the architecture. Synthesis results show that the architecture has high throughput/ slice value compared with state-of-the-art SHA-256 implementations. The article also promulgates a comparison between high-level synthesis and RTL design.

Sign in / Sign up

Export Citation Format

Share Document