• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.14 no.4
• /
• pp.314-322
• /
• 2021

Skeleton-based action recognition has attracted considerable attention in human action recognition. Recent methods for skeleton-based action recognition employ spatiotemporal graph convolutional networks (GCNs) and have remarkable performance. However, most of them have heavy computational complexity for robust action recognition. To solve this problem, we propose a shuffle graph convolutional network (SGCN) which is a lightweight graph convolutional network using pointwise group convolution rather than pointwise convolution to reduce computational cost. Our SGCN is composed of spatial and temporal GCN. The spatial shuffle GCN contains pointwise group convolution and part shuffle module which enhances local and global information between correlated joints. In addition, the temporal shuffle GCN contains depthwise convolution to maintain a large receptive field. Our model achieves comparable performance with lowest computational cost and exceeds the performance of baseline at 0.3% and 1.2% on NTU RGB+D and NTU RGB+D 120 datasets, respectively.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.4
• /
• pp.1851-1863
• /
• 2015

– In the paper, to handle the problem of traditional bag-of-features model ignoring the spatial relationship of local features in human action recognition, we proposed a Multiscale Spatial Position Coding under Locality Constraint method. Specifically, to describe this spatial relationship, we proposed a mixed feature combining motion feature and multi-spatial-scale configuration. To utilize temporal information between features, sub spatial-temporal-volumes are built. Next, the pooled features of sub-STVs are obtained via max-pooling method. In classification stage, the Locality-Constrained Group Sparse Representation is adopted to utilize the intrinsic group information of the sub-STV features. The experimental results on the KTH, Weizmann, and UCF sports datasets show that our action recognition system outperforms the classical local ST feature-based recognition systems published recently.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.327-338
• /
• 2022

Classification models for human action recognition require robust features and large training sets for good generalization. However, data augmentation methods are employed for imbalanced training sets to achieve higher accuracy. These samples generated using data augmentation only reflect existing samples within the training set, their feature representations are less diverse and hence, contribute to less precise classification. This paper presents new data augmentation and action representation approaches to grow training sets. The proposed approach is based on two fundamental concepts: virtual video generation for augmentation and representation of the action videos through robust features. Virtual videos are generated from the motion history templates of action videos, which are convolved using a convolutional neural network, to generate deep features. Furthermore, by observing an objective function of the genetic algorithm, the spatiotemporal features of different samples are combined, to generate the representations of the virtual videos and then classified through an extreme learning machine classifier on MuHAVi-Uncut, iXMAS, and IAVID-1 datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.5
• /
• pp.2327-2347
• /
• 2018

Human actions can be recognized from depth sequences. In the proposed algorithm, we initially construct depth, motion maps (DMM) by projecting each depth frame onto three orthogonal Cartesian planes and add the motion energy for each view. The body part of the action (BPoA) is calculated by using bounding box with an optimal window size based on maximum spatial and temporal changes for each DMM. Furthermore, feature vector is constructed by using BPoA for each human action view. In this paper, we employed an ensemble based learning approach called Rotation Forest to recognize different actions Experimental results show that proposed method has significantly outperforms the state-of-the-art methods on Microsoft Research (MSR) Action 3D and MSR DailyActivity3D dataset.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.513-515
• /
• 2023

본 논문에서는 그리드 탐색과 트랜스포머를 사용한 그룹 행동 인식 모델을 제안한다. 추출된 여러 사람의 스켈레톤 정보를 차분 벡터, 변위 벡터, 관계 벡터로 변환하고 사람별로 묶어 이를 TimeDistributed 함수에 넣고 풀링을 한다. 이를 트랜스포머 모델의 입력으로 넣고 그룹 행동 인식 분류를 출력하였다. 논문에서 3가지 벡터를 입력으로 하여 합치고 트랜스포머 계층을 거친 모델과 3가지 벡터를 입력으로 하고 계층적으로 트랜스포머 모델을 거쳐 행동 인식 분류를 출력하는 두 가지 모델을 제안한다. 3가지 벡터를 합친 모델에서 클래스 분류 정확도는 CAD 데이터 세트 96.6%, Volleyball 데이터 세트 91.4%, 계층적 트랜스포머 모델은 CAD 데이터 세트 96.8%, Volleyball 데이터 세트 91.1%를 얻었다

• ETRI Journal
• /
• v.37 no.6
• /
• pp.1199-1210
• /
• 2015

Automatic emotion recognition based on facial cues, such as facial action units (AUs), has received huge attention in the last decade due to its wide variety of applications. Current computer-based automated two-phase facial emotion recognition procedures first detect AUs from input images and then infer target emotions from the detected AUs. However, more robust AU detection and AU-to-emotion mapping methods are required to deal with the error accumulation problem inherent in the multiphase scheme. Motivated by our key observation that a single AU detector does not perform equally well for all AUs, we propose a novel two-phase facial emotion recognition framework, where the presence of AUs is detected by group decisions of multiple AU detectors and a target emotion is inferred from the combined AU detection decisions. Our emotion recognition framework consists of three major components - multiple AU detection, AU detection fusion, and AU-to-emotion mapping. The experimental results on two real-world face databases demonstrate an improved performance over the previous two-phase method using a single AU detector in terms of both AU detection accuracy and correct emotion recognition rate.

• Korean Journal of Family Social Work
• /
• no.62
• /
• pp.103-139
• /
• 2018

The purpose of this study was to explore the social workers' reflections developed during the participatory action research seeking practice methods in the community welfare center single parent self-help groups. Nine social workers and researchers sharing difficulties in the community welfare center single parent self-help group practice were involved in the participatory action research. We performed a circulatory research repeating 'plan-action-reflection-replan' cycles through ten sessions of research meetings including peer supervision in parallel with 4 self-help group practices. Multiple practice methods were derived as the result among which change in the social workers' perception was the key. We focused on the result that the social workers' perception was changed by reflections from participatory action research and the contents of the reflections were qualitatively analyzed using materials acquired during the study. As a result the following 5 themes were derived; 'recognition and worry on the lack of appropriate practice for the self-help group', 'specification of the social workers' role based on the changed perspective on the member-social worker relation', 'confirmation of the possibility of the concerned person initiated practice', 'detailed understanding of the single parent self-help group in the community welfare center', and 'renewed recognition of the development of the single parent self-help group in the community welfare center'. Based on these results, empowerment and reflection/reflective practice for single parent self-help group in the community welfare center were discussed.

• Journal of Nutrition and Health
• /
• v.43 no.6
• /
• pp.597-606
• /
• 2010

This study investigates the change of level of psychosocial factors of decisional balance, processes of change, and self-efficacy after nutrition education on vegetable and fruit intake of children in pre-action stages. Forty-seven 5th grade elementary school children in Yeongi-Gun, Chungnam-Do were categorized into pre-action stages from a total of 71 children by four-stage questions, and then finally into two stages of precontemplation and contemplation & preparation. After 6 weeks nutrition education, recognition of both of pros and cons of vegetable intake were improved in the two groups of precontemplation and contemplation & preparation. In fruits intake, however, recognition of pros was not changed after nutrition education in either stage, but recognition of cons was improved only in the precontemplation stage. However, no difference in any psychosocial variables was found between precontemplation stage and contemplation & preparation stage. After the nutrition education, the score of consciousness raising of vegetable intake belonging to cognitive process of change increased in the subjects of the precontemplation stage, wheres the scores of dramatic relief, self-reevaluation, environment reevaluation and social liberation increased in the group of contemplation & preparation. In fruit intake, however, no change was found except in the environment reevaluation in the precontemplation group. The behavioral process of change of vegetable and fruit intake did not show any change in either group after nutrition education. Self-efficacy scores of vegetable intake of keeping the foods at hand, eating the recommended number of servings and making time to eat increased in the precontemplation group, while in the contemplation & preparation group, self efficacy scores of shopping for a variety of vegetables, instead of keeping the foods at hands, as well as the scores of the rest two items of eating the recommended number of servings and making time to eat increased after nutrition education. However, in case of fruits intake, no difference was shown in the precontemplation stage except an item of keeping the foods at hand. The results indicated that the stage of change was progressed in the pre-action stage after nutrition education and the scores of many psychosocial factors increased in both pre-action stages. It is suggested that the recognition of cons as well as the cognitive process of change and self-efficacy could be best used as indicators of the assessment of nutrition education program for vegetable intake.

• Journal of Family Resource Management and Policy Review
• /
• v.13 no.3
• /
• pp.181-205
• /
• 2009

The purpose of this study is to investigate some Influencing factors related to the wealth perception and property increase of 519 women in Seoul. The research was conducted from the end of February 15 to the beginning of January 20, 2009. The figures obtained were analyzed by frequency, percentage, mean, x-test, t-test, F-test, chi-square, and Duncan's F-test with SPSS pc+ The major findings of who study are summarized as follows: l.There were significant differences between the wealth recognition (economic plan affirmative and negative perception of wealth) for women according to social and psychological variables; economic plans have a higher value for married women, older women, those who have followed a postgraduate university course, those with higher income and expenditure, higher life satisfaction, and high and middle satisfaction with their economic life. In particular, economic plans varied significantly according to expectation of wealth in old age, and future economic prospects in the middle group. 2. Increased wealth and economic according to social and psychological variables (marital status, age, education, income, expenditure, life satisfaction, satisfaction with economic life, expectation of wealth in old age, future economic prospects) were shown to be associated with the following methods of increasing property: use of savings banks (p<.001) showed a rate of deposit of 20%-40% (p<.001) saving and thrift was a practice (p<001) that was related to the cost of living; and education expenditure (p<.001) as a property preparation period was exhibited in women aged 25-30 (p<.001). 3. There were significant differences in the recognition of wealth according to strategies for increasing wealth. There were no significant differences between mean of property acquisition and point of expenditure or the property preparation period in the perception of wealth. The related variables in rate of deposits were under 20% according to economic plan and negative wealth perception. In contrast, in the middle and high group, the economic plan and negative wealth perception were 20%-40% in relation to the rate of deposits. Variables related to the action of saving and thrift and property acquisition were practiced in the low and middle group of affirmative wealth recognition and the high group of negative wealth perception. The high group for wealthrecognition and the low group for wealthnegation exhibited little savings, thrift, and property accumulation. Those variables which point an expenditure the cost of living were education expenditure in all groups of negative wealth perception. The women in the study practiced more soundness economic planning according to their increased wealth and economic strategy. This study suggests that women should have positive wealth perception in their lives.