Joint Coordinates Research Articles

Angle based hand gesture recognition using graph convolutional network

AbstractHand gesture recognition has attracted huge interest in the areas of autonomous driving, human computer systems, gaming and many others. Skeleton based techniques along with graph convolutional networks (GCNs) are being popularly used in this field due to the easy estimation of joint coordinates and better representation capability of graphs. Simple hand skeleton graphs are unable to capture the finer details and complex spatial features of hand gestures. To address these challenges, this work proposes an “angle‐based hand gesture graph convolutional network” (AHG‐GCN). This model introduces two additional types of novel edges in the graph to connect the wrist with each fingertip and finger's base, explicitly capturing their relationship, which plays an important role in differentiating gestures. Besides, novel features for each skeleton joint are designed using the angles formed with fingertip/finger‐base joints and the distance among them to extract semantic correlation and tackle the overfitting problem. Thus, an enhanced set of 25 features for each joint is obtained using these novel techniques. The proposed model achieves 90% and 88% accuracy for 14 and 28 gesture configurations for the DHG 14/28 dataset and, 94.05% and 89.4% accuracy for 14 and 28 gesture configurations for the SHREC 2017 dataset, respectively.

Analysis of inter-joint coordination during the sit-to-stand and stand-to-sit tasks in stroke patients with hemiplegia

BackgroundInter-joint coordination is an important factor affecting postural stability, and its variability increases after fatigue. This study aimed to investigate the coordination pattern of lower limb joints during the sit-to-stand (Si-St) and stand-to-sit (St-Si) tasks in stroke patients and explore the influence of duration on inter-joint coordination.MethodsThirteen stroke hemiplegia patients (five with left paretic and eight right paretic) and thirteen age-matched healthy subjects were recruited. The Si-St and St-Si tasks were performed while each subject’s joint kinematics were recorded using a three-dimensional motion capture system. Sagittal joint angles of the bilateral hip, knee and ankle joints as well as the movement duration were extracted. The angle-angle diagrams for the hip-knee, hip-ankle and knee-ankle joint were plotted to assess the inter-joint coordination. The inter-joint coordination was quantified using geometric characteristics of the angle-angle diagrams, including perimeter, area and dimensionless ratio. The coefficient of variation (CV) was performed to compare variability of the coordination parameters.ResultsThere were no significant differences in the perimeter, area and dimensionless ratio values of the bilateral hip-knee, hip-ankle and knee-ankle inter-joints during Si-St and St-Si tasks in the stroke group. The perimeter values of bilateral hip-knee and knee-ankle inter-joints in the stroke group were lower (P<0.05) than in the healthy group during Si-St and St-Si tasks. Although no significant bilateral differences were found, the inter-joint coordination in stroke patients decreased with the increased movement duration of both Si-St and St-Si tasks. Additionally, the CV of the hip-knee inter-joint area during the Si-St task in the stroke group was less than (P<0.05) that in the healthy group.ConclusionStroke patients exhibit different inter-joint coordination patterns than healthy controls during the Si-St and St-Si tasks. The duration affects joint coordination, and inter-joint coordination is limited on the hemiplegic side joint pairs, which may lead to inconsistency in the rhythm of the left and right leg inter-joint movements and increase the risk of falls. These findings provide new insights into motor control rehabilitation strategies and may help planning targeted interventions for stoke patients with hemiplegia.

Mobile Gait Analysis

This project aims to develop an extension to automated gait analysis that makes gait analysis available on smart devices. The alternative may serve as a baseline for future implementations that are cheaper, user-friendly and accessible to an ordinary smartphone or web browser. Accessibility of gait analysis on an application encourages people to check their walking patterns more regularly, and if the issue is very severe, they can take the next step of contacting a specialist. By collaborating with the Podiatry Department of the University of Malta and the Chinese Academy of Sciences Institute of Automation (CASIA), a considerable amount of gait data was acquired. The data consists of videos of people walking regularly or irregularly. But videos are not enough for the development of our system. The videos were inputted into a pose estimator whose goal was to outline the skeleton of the person throughout the video. Additionally, the pose estimator was modified to record the coordinates of the main joints concerning a gait cycle (hip, knee and ankle). These coordinates were then plotted as a scatter plot where the gait cycle is generated. With the coordinates extracted, kinematics were also extracted to create another model which detects different features for gait analysis. After the gait cycle of each video was extracted, the next step was to classify whether that gait cycle was either regular or irregular. This goal is achieved by passing the extracted data through the VGG16 architecture. The application was tested out on people which have either bad, good or slightly bad gaits to investigate the rigidity of the system. After a series of experiments, it can be concluded that the system performs with 94% accuracy just by using a mobile phone.

Hybrid Administrative Interfaces: Authority Delegation and Reversion in Strategic Alliances

Steering committees are pivotal for governing complex collaborations by consensus to facilitate coordination and knowledge sharing. Although consensus-based governance promotes mutuality, it can also cause deadlocks, stalling expeditious decision making. We examine the conditions under which alliance partners delegate decision-making authority to steering committees as well as the conditions under which authority over discordant matters can be relocated to one of the alliance partners. We argue that joint coordination concerns increase the likelihood of authority delegation, whereas the higher costs and stakes associated with decision stalemates provide grounds for authority reversion. Empirical analyses of strategic alliances in the biopharmaceutical industry support our arguments. Our paper demonstrates the versatility of contractually defined administrative interfaces in alliance governance, allowing partners to coordinate bilaterally and adapt hierarchically as and when required.

An Extended Labanotation Generation Method Based on 3D Human Pose Estimation for Intangible Cultural Heritage Dance Videos

To address the issues of low accuracy in existing 3D human pose estimation (HPE) methods and the limited level of details in Labanotation, we propose an extended Labanotation generation method for intangible cultural heritage dance videos based on 3D HPE. First, a 2D human pose sequence of the performer is inputted along with spatial location embeddings, where multiple spatial transformer modules are employed to extract spatial features of human joints and generate cross-joint multiple hypotheses. Afterward, temporal features are extracted by a self-attentive module and the correlation between different hypotheses is learned using bilinear pooling. Finally, the 3D joint coordinates of the performer are predicted, which are matched with the corresponding extended Labanotation symbols using the Laban template matching method to generate extended Labanotation. Experimental results show that, compared with VideoPose and CrossFormer algorithms, the Mean Per Joint Position Error (MPJPE) of the proposed method is reduced by 3.7[Formula: see text]mm and 0.6[Formula: see text]mm, respectively on Human3.6M dataset, and the generated extended Labanotation can better describe the movement details compared with the basic Labanotation.

Virtual Reality for Action Observation Does Not Enhance Motor Learning of a Complex Weightlifting Task

Action observation, where a learner observes a model demonstrating a skill, has long been established in the literature as an effective means to acquire motor skills. Developments in virtual reality technology have made it possible for a 3D action observation viewing perspective, theoretically providing a viewing experience similar to observing a live performer. However, very little work has compared these two media and their effects on motor learning outcomes. In this current study, healthy novice participants to Olympic lifting (specifically the clean and jerk; n = 36) learned the exercise through observing a model demonstration in virtual reality (3D group) or on a computer screen (2D group). A third group (control) did not engage in action observation. Results indicate that following training, the 2D group more frequently used a three-step lifting pattern, and the 2D and 3D groups lifted with a significantly greater horizontal displacement compared with the control group. Also, the 2D group was more likely to use a proximal-to-distal joint coordination strategy during the second pull of the exercise. These results further the idea that only small parts of pertinent visual information are needed for action observation, and that virtual reality may provide too much information to novice learners.

Using One Health assessments to leverage endemic disease frameworks for emerging zoonotic disease threats in Libya.

Continued emergence, re-emergence and spread of zoonotic diseases demonstrates the imperative need for multisectoral communication and joint coordination of disease detection and response. While there are existing international frameworks underpinning One Health capacity building for pandemic prevention and response, often guidance does not account for challenges faced by countries undergoing long-term conflict and sociopolitical instability. The purpose of this research was to identify Libya's laboratory and surveillance networks and routes of inter- and multisectoral communication and coordination for priority zoonotic diseases. The One Health Systems Assessment for Priority Zoonoses (OH-SAPZ) tool is an established methodology that was adapted and applied to the Libyan context to support prioritization of zoonotic diseases, development of systems map schematics outlining networks of communication and coordination, and analysis of operations for targeted capacity building efforts. Five zoonotic diseases were selected to undergo assessment: highly pathogenic avian influenza, brucellosis, Rift Valley fever, leishmaniasis and rabies. Through decisive acknowledgement of Libya's unique health setting, we mapped how patient and sample information is both communicated within and between the human, animal and environmental health sectors, spanning from local index case identification to international notification. Through our assessment we found strong communication within the public and animal health sectors, as well as existing multisectoral coordination on zoonotic disease response. However, local-level communication between the sectors is currently lacking. Due to the ongoing conflict, resources (financial and human) and access have been severely impacted, resulting in limited laboratory diagnostic capacity and discontinued disease prevention and control measures. We sought to identify opportunities to leverage existing operations for endemic diseases like brucellosis for emerging zoonotic threats, such as Rift Valley fever. Analysis of these operations and capabilities supports the development of targeted recommendations that address gaps and may be used as an implementation guide for future One Health capacity building efforts.

CHAN: Skeleton based action recognition by multi‐level feature learning

AbstractSkeleton‐based action recognition has been continuously and intensively studied. However, dynamic 3D skeleton data are difficult to be popularized in practical applications due to the restricted data acquisition conditions. Although the action recognition method based on 2D pose information extracted from RGB video can effectively avoid the influence of complex background, it is susceptible to factors such as video jitter and joint overlap. To reduce the interference of the aforementioned factors, we use two‐dimensional skeletal joint coordinate modal information to represent the changes in human body posture. First, we use a target detector and pose estimation algorithm to obtain the joint coordinates of each frame sample from RGB video. Then the feature extraction network is combined to perform multi‐level feature learning to establish correspondence between actions and corresponding multi‐level features. Finally, the hierarchical attention mechanism is introduced to design the model named CHAN. By calculating the association between elements, the weight of the action classification is redistributed. Extensive experiments on three datasets demonstrate the effectiveness of our proposed method.

GaitGCN++: Improving GCN-based gait recognition with part-wise attention and DropGraph

Gait recognition is becoming one of the promising methods for biometric authentication owing to its self-effacing nature. Contemporary approaches of joint position-based gait recognition generally model gait features using spatio-temporal graphs which are often prone to overfitting. To incorporate long-range relationships among joints, these methods utilize multi-scale operators. However, they fail to provide equal importance to all joint combinations resulting in an incomplete realization of long-range relationships between joints and important body parts. Furthermore, only considering joint coordinates may fail to capture discriminatory information provided by the bone structures and motion. In this work, a novel multi-scale graph convolution approach, namely ‘GaitGCN++’, is proposed, which utilizes joint and bone information from individual frames and joint-motion data from consecutive frames providing a comprehensive understanding of gait. An efficient hop-extraction technique is utilized to understand the relationship between closer and further joints while avoiding redundant dependencies. Additionally, traditional graph convolution is enhanced by leveraging the ‘DropGraph’ regularization technique to avoid overfitting and the ‘Part-wise Attention’ to identify the most important body parts over the gait sequence. On the benchmark gait recognition dataset CASIA-B and GREW, we outperform the state-of-the-art in diversified and challenging scenarios.

Spatial Hard Attention Modeling via Deep Reinforcement Learning for Skeleton-Based Human Activity Recognition

Deep learning-based algorithms have been very successful in skeleton-based human activity recognition. Skeleton data contains 2-D or 3-D coordinates of human body joints. The main focus of most of the existing skeleton-based activity recognition methods is on designing new deep architectures to learn discriminative features, where all body joints are considered equally important in recognition. However, the importance of joints varies as an activity proceeds within a video and across different activities. In this work, we hypothesize that selecting relevant joints, prior to recognition, can enhance performance of the existing deep learning-based recognition models. We propose a spatial hard attention finding method that aims to remove the uninformative and/or misleading joints at each frame. We formulate the joint selection problem as a Markov decision process and employ deep reinforcement learning to train the proposed spatial-attention-aware agent. No extra labels are needed for the agent’s training. The agent takes a sequence of features extracted from skeleton video as input and outputs a sequence of probabilities for joints. The proposed method can be considered as a general framework that can be integrated with the existing skeleton-based activity recognition methods for performance improvement purposes. We obtain very competitive activity recognition results on three commonly used human activity recognition datasets.

Learning from intersectoral initiatives to respond to the needs of refugees, asylum seekers, and migrants without status in the context of COVID-19 in Quebec and Ontario: a qualitative multiple case study protocol

BackgroundRefugees, asylum seekers, and migrants without status experience precarious living and working conditions that disproportionately expose them to coronavirus disease 2019 (COVID-19). In the two most populous Canadian provinces (Quebec and Ontario), to reduce the vulnerability factors experienced by the most marginalized migrants, the public and community sectors engage in joint coordination efforts called intersectoral collaboration. This collaboration ensures holistic care provisioning, inclusive of psychosocial support, assistance to address food security, and educational and employment assistance. This research project explores how community and public sectors collaborated on intersectoral initiatives during the COVID-19 pandemic to support refugees, asylum seekers, and migrants without status in the cities of Montreal, Sherbrooke, and Toronto, and generates lessons for a sustainable response to the heterogeneous needs of these migrants.MethodsThis theory-informed participatory research is co-created with socioculturally diverse research partners (refugees, asylum seekers and migrants without status, employees of community organizations, and employees of public organizations). We will utilize Mirzoev and Kane’s framework on health systems’ responsiveness to guide the four phases of a qualitative multiple case study (a case being an intersectoral initiative). These phases will include (1) building an inventory of intersectoral initiatives developed during the pandemic, (2) organizing a deliberative workshop with representatives of the study population, community, and public sector respondents to select and validate the intersectoral initiatives, (3) interviews (n = 80) with community and public sector frontline workers and managers, municipal/regional/provincial policymakers, and employees of philanthropic foundations, and (4) focus groups (n = 80) with refugees, asylum seekers, and migrants without status. Qualitative data will be analyzed using thematic analysis. The findings will be used to develop discussion forums to spur cross-learning among service providers.DiscussionThis research will highlight the experiences of community and public organizations in their ability to offer responsive services for refugees, asylum seekers, and migrants without status in the context of a pandemic. We will draw lessons learnt from the promising practices developed in the context of COVID-19, to improve services beyond times of crisis. Lastly, we will reflect upon our participatory approach—particularly in relation to the engagement of refugees and asylum seekers in the governance of our research.

Lower extremity coordination strategies to mitigate dynamic knee valgus during landing in males and females

Frontal and sagittal plane landing biomechanics differ between sexes but reported values don’t account for simultaneous segment or joint motion necessary for a coordinated landing. Frontal and sagittal plane coordination patterns, angles, and moments were compared between 28 males and 28 females throughout the landing phase of a drop vertical jump. Females landed with less isolated thigh abduction (p = 0.018), more in-phase motion (p < 0.001), and more isolated shank adduction (p = 0.028) between the thigh and shank in the frontal plane compared with males. Females landed with less in-phase (p = 0.012) and more anti-phase motion (p = 0.019) between the thigh and shank in the sagittal plane compared with males. Females landed with less isolated knee flexion (p = 0.001) and more anti-phase motion (p < 0.001) between the sagittal and frontal plane knee coupling compared with males. Waveform and discrete metric analyses revealed females land with less thigh abduction from 20 % to 100 % and more shank abduction from 0 to 100 % of landing, smaller knee adduction at initial contact (p = 0.002), greater peak knee abduction angles (p = 0.015), smaller knee flexion angles at initial contact (p = 0.035) and peak (p = 0.034), greater peak knee abduction moments (p = 0.024), greater knee abduction angles from 0 to 13 % and 19 to 30 %, greater knee abduction moments from 19 to 25 %, and smaller knee flexion moments from 3 to 5 % of landing compared with males. Females utilize greater frontal plane motion compared with males, which may be due to different inter-segmental joint coordination and smaller sagittal plane angles. Larger knee abduction angles and greater knee adduction motion in females are due to aberrant shank abduction rather than thigh adduction.

Joint Coordination With a Change in Task Constraint During Accurate Overhead Throwing

In sports situations, players may be required to throw at different speeds. The question of how skilled players throw the ball accurately to the desired location under different speed conditions is of interest to biomechanics researchers. Previous research suggested that throwers use different types of joint coordination. However, joint coordination with a change in throwing speed has not been studied. Here, we show the effects of changes in throwing speed on joint coordination during accurate overhead throwing. Participants were seated on a low chair with their trunk fixed and threw a baseball aimed at a target under 2 different speed conditions (slow and fast). In the slow condition, the elbow flexion/extension angle coordinated with other joint angles and angular velocities to reduce the variability of the vertical hand velocity. In the fast condition, the shoulder internal/external rotation angle and the shoulder horizontal flexion/extension angular velocity coordinated with other joint angles and angular velocities to reduce the variability of the vertical hand velocity. These results showed that joint coordination differed with changes in throwing speed, indicating that joint coordination is not always fixed, but may differ depending on the task constraints, such as throwing speed.

On Collaboration of Local Actors With Security Force Assistance in Missions and Non-Article 5 Crisis Response Operations

Abstract This paper examines aspects of the coordination and cooperation of local actors in Security Force Assistance (SFA) missions, and in particular in Non-Article 5 Crisis Response Operations (NA5CRO). The process of building relationships with local actors and coordinating efforts with them to achieve mission success is discussed as well. This process is based on the understanding that local actors have critical knowledge of the operational environment and can provide valuable insights that enhance mission effectiveness. The need to understand the operational environment and build effective relationships with local security forces as well as other stakeholders is also addressed. The article explains how coordination and cooperation with local actors are essential to achieving mission success and can improve the effectiveness of the SFA in their participation in crisis response operations. In summary, the paper emphasizes the critical importance of coordination and cooperation with local actors in SFA missions and into Humanitarian Assistance Operations that are part of NA5CRO. This approach can enhance the effectiveness of the mission, increase situational awareness, and reduce the risk of mission failure. It requires a commitment to building effective relationships with local security forces and other stakeholders, joint coordination among all actors involved in the mission, and cultural sensitivity and awareness in the event of a crisis.

Inter-joint coordination variability is associated with pain severity and joint loading in persons with knee osteoarthritis.

As the lower extremity is a linked-joint system, the contribution of movements at the hip and ankle, in addition to the knee, to gait patterns should be considered for persons with knee osteoarthritis (OA).However, the relationships of joint coordination variability to OA symptoms, particularly knee pain, and joint loading is unknown. The purpose of this study was to determine the relationship of joint coordination variability to knee pain severity and joint loading in persons with knee OA. Thirty-fourparticipants with knee OA underwent gait analysis. Vector coding was used to assess coordination variability during the early, mid, and late stance phase. Hip-knee coupling angle variability (CAV)during midstance was associated with Knee Injury and Osteoarthritis Outcome Score (KOOS)pain (r = -0.50, p = 0.002) and Visual Analog Scalepain (r = 0.36, p = 0.04). Knee-ankle CAV during midstance was associated with KOOS pain (r = -0.34, p = 0.05). Hip-knee CAV during early and midstance were associated with knee flexion moment (KFM)impulses (r = -0.46, p = 0.01). Knee-ankle CAV during early and midstance were associated with peak KFM (r = -0.51, p < 0.01; r = -0.70, p < 0.01). Moreover, knee-ankle CAV during early, mid, and late stance phase were associated with KFM impulses (r = -0.53, p < 0.01; r = -0.70, p < 0.01; r = -0.54, p < 0.01). These findings suggest that joint coordination variability may be a factor that influences pain and knee joint loading in persons with knee OA. Statement of Clinical Significance: Movement coordination of the hip, knee, and ankle should be considered in the clinical management and future research related to knee OA.

Attentional focus effects on joint covariation in a reaching task

Adopting an external focus of attention (EF) has been found beneficial over internal focus (IF) for performing motor skills. Previous studies primarily examined focus of attention (FOA) effects on performance outcomes (such as error and accuracy), with relatively less emphasis on movement coordination. Given that human movements are kinematically and kinetically abundant (Gefland & Latash, 1998), FOA instructions may change how motor abundance is utilized by the CNS. This study applied the uncontrolled manifold analysis (UCM) to address this question in a reaching task. Healthy young adults (N = 38; 22 ± 1 yr; 7 men, 31 women) performed planar reaching movements to a target using either the dominant or nondominant arm under two different FOA instructions: EF and IF. Reaching was performed without online visual feedback and at a preferred pace. Joint angles of the clavicle-scapula, shoulder, elbow, and wrist were recorded, and their covariation for controlling dowel endpoint position was analyzed via UCM. As expected, IF led to a higher mean radial error than EF, driven by increases in aiming bias and variability. Consistent with this result, the UCM analysis showed that IF led to higher goal-relevant variance among the joints (VORT) compared to EF starting from the first 20% of the reach to the end. However, the goal-irrelevant variance (VUCM)—index of joint variance that does not affect the end-effector position—did not show FOA effects. The index of stability of joint coordination with respect to endpoint position (ΔV) was also not different between the EF and IF. Consistent with the constrained action hypothesis, these results provide evidence that IF disrupted goal-relevant joint covariation starting in the early phases of the reach without affecting goal-irrelevant coordination.

Equilibrium analysis for prosumer’s participation in joint electricity and carbon markets

The current electricity system has a large number of carbon emissions. As an environmentally friendly and low-carbon electricity generation source, distributed energy sources gradually get more and more attention. However, the participation of a large number of distributed energy sources brings uncertainty to the operation of the electricity market. Prosumers could effectively avoid the instability caused by the participation of distributed energy sources in the electricity market. At present, the strategy of the prosumer participating in the electricity market has been thoroughly studied. However, equilibrium analysis for prosumer’s participation in joint electricity and carbon markets still needs to be studied in depth. Therefore, we propose a method for the prosumer to participate in the joint electricity and carbon markets, which could effectively reduce carbon emissions. First, we build a carbon market model. Then, the equilibrium model of the prosumer’s participation in joint electricity and carbon markets is constructed. The equilibrium model can realize the equilibrium between the coordination of prosumers and joint markets. Finally, we adopt an iterative algorithm to solve the equilibrium model. Prosumers and joint electricity and carbon markets transmit marginal information to each other. According to the marginal information. The equilibrium between Prosumer and joint markets is solved through iteration by self-optimization and self-decision. Case studies based on the IEEE 30-bus system verify the effectiveness of the model and method proposed in this paper.

Stacked graph bone region U-net with bone representation for hand pose estimation and semi-supervised training

3D hand estimation from 2D joint information is an essential task in human-machine interaction, which has achieved great progress as an application of deep learning. However, regression-based methods do not perform well because the structural information is not effectively exploited, and the joint coordinates are variable. To address these issues, the hand pose is represented with bone vectors instead of joint coordinates in this study, which are stabler to learn and allow for easier encoding of the hand geometric structure and joint dependency. A novel graph bone region U-Net is specifically designed for bone representation to learn multiscale structural features, where the proposed novel elements (graph convolution, pooling and unpooling) incorporate hand structural knowledge. Under the introduced “finger-to-hand” framework, the network gradually decreases the scale from bone to finger to hand for learning more meaningful multiscale features. Moreover, the unit network is stacked repeatedly to extract multilevel features. Based on the above network, a simple but effective semi-supervised approach is introduced to address the lack of 3D hand pose labels. Many experiments are conducted to evaluate the proposed approach on two challenging datasets. The experimental results show that the proposed supervised approach outperforms the state-of-the-art methods, and the proposed semi-supervised approach can still achieve favorable performance when the labeled data are scarce.

Application of Wearable Gloves for Assisted Learning of Sign Language Using Artificial Neural Networks

This study proposes the design and application of wearable gloves that can recognize sign language expressions from input images via long short-term memory (LSTM) network models and can learn sign language through finger movement generation and vibration motor feedback. It is difficult for nondisabled people who do not know sign language to express sign language accurately. Therefore, we suggest the use of wearable gloves for sign language education to help nondisabled people learn and accurately express sign language. The wearable glove consists of a direct current motor, a link (finger exoskeleton) that can generate finger movements, and a flexible sensor that recognizes the degree of finger bending. When the coordinates of the hand move in the input image, the sign language motion is fed back through the vibration motor attached to the wrist. The proposed wearable glove can learn 20 Korean sign language words, and the data used for learning are configured to represent the joint coordinates and joint angles of both the hands and body for these 20 sign language words. Prototypes were produced based on the design, and it was confirmed that the angle of each finger could be adjusted. Through experiments, a sign language recognition model was selected, and the validity of the proposed method was confirmed by comparing the generated learning results with the data sequence. Finally, we compared and verified the accuracy and learning loss using a recurrent neural network and confirmed that the test results of the LSTM model showed an accuracy of 85%.

Establishment and Activities of the Black Sea Ukrainian Community in the City Of Sevastopol in 1917

This study, dedicated to the activities of the Black Sea Ukrainian Community of the city of Sevastopol, is an attempt to find out and organize the chronology and factology of the activities of one of the leading organizations of the military-civilian Ukrainian movement in Sevastopol in 1917.Occasionally, little-known data on personalities associated with the specified organization and contemporary events is provided. The research used data from the Ukrainian press for the specified period and the memoirs of participants and eyewitnesses of the events. It is noted that the basis of the organization was the illegal circle “Kobzar”, which operated in Sevastopol in 1900–1917 and was formed mainly from officers of the Black Sea Fleet and representatives of the intelligentsia, in particular, educators. The change in the organizational form was caused by the revolutionary events in the Russian Empire in March 1917 and the onset of a new phase of the Ukrainian national liberation movement. The largest number of members of the community, which acted as a coordination center for Ukrainian formations on ships and in coastal parts of the Black Sea Fleet, were officers and sailors, but the organization also aimed primarily at educational work with the aim of raising awareness among the masses, continuing the work of the “Kobzar” circle. At the beginning of the community activity, the bearers of the idea of Ukrainianization of the army and navy were mainly a few representatives of the youth, who later formed an opposition to the older members of the community but acted in a constructive direction; later, the community adopted a resolution calling for the Ukrainianization of the Black Sea Fleet. Also, the Black Sea Ukrainian community of the city of Sevastopol established contacts with representatives of the Ukrainian authorities for the purpose of joint coordination of actions. The organization stopped its main activities approximately at the end of 1917 due to the Red Terror in Sevastopol, but approximately in the 1920s, “The Union of Members of the Black Sea Ukrainian Community of the City of Sevastopol in Emigration” was formed in the Czechoslovak Republic.