Batch Entropy Supervised Convolutional Neural Networks for Feature Extraction and Harmonizing for Action Recognition

Hossain, Imtiaz; Siddique, Ashraf; Hossain, Alamgir; Hossain, Delowar; Huh, Eui‐Nam

doi:10.1109/access.2020.3037529

Cited by 11 publications

(12 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, the integrated information is re-integrated with the next video frame to increase the training length by iterative methods, thus increasing the training and recognition accuracy. Hossain et al (2020) learned the heterogeneous spatiotemporal cues of video motion recognition. They performed an in-depth study of feature maps, focusing on extracting features with highly discriminative rows.…”

Section: Deep Learning For Human Motion Recognitionmentioning

confidence: 99%

The Application of Multimedia and Deep Learning in the Integration of Professional and Innovative Education in Colleges

2023

International Journal of Information Technologies and Systems Approach

View full text Add to dashboard Cite

In this paper, the authors study the assessment of professional practice courses in the integrated education of specialization and innovation, and use the multimedia and deep learning technology to complete the action recognition of students in practice courses. Firstly, the skeletal features are extracted from multimedia video data by Openpose algorithm, which is used for subsequent classification while ensuring privacy; then the LSTM method is used to recognize typical motions in student practice, and the average recognition result exceeds 89%; finally, practical application tests are conducted for laboratory and office scenes, and the results illustrate that the proposed framework performs well in the tests with recognition rate exceeding 80%. The algorithm framework provides a new idea for the curriculum setting and evaluation method of professional practice education, and gives data guarantee for their integration and innovation education.

show abstract

Section: Deep Learning For Human Motion Recognitionmentioning

confidence: 99%

The Application of Multimedia and Deep Learning in the Integration of Professional and Innovative Education in Colleges

2023

International Journal of Information Technologies and Systems Approach

View full text Add to dashboard Cite

show abstract

“…The regional approach is applied through an assessment and treatment model (Extended data 1 32 ). The assessment model screens CR as the 1 st step of problems and examine the whole region of cervical, thoracic and upper limb region as 2 nd and 3 rd step.…”

Section: Mckenzie Mechanical Diagnosis Of Therapy (Mdt)mentioning

confidence: 99%

Physiotherapy versus pain medication for managing chronic cervical radiculopathy: protocol for a multi-arm parallel-group randomized clinical trial

et al. 2023

View full text Add to dashboard Cite

Background: Cervical radiculopathy (CR) is one of the prevalent causes of neck pain and disability. Physiotherapy and pain medications are the common nonoperative management, and in physiotherapy, there are many concepts of assessment and management. This study aims to determine the comparative effectiveness of three specialized physiotherapy approaches or only pain medications for managing CR cases. Methods: A prospective, assessor, and participant-blind, four-arm randomized control trial (RCT) has been planned to conduct on 160 patients with chronic cervical radiculopathy in 4 specialized centers of Dhaka city recruited between July and December 2022. Four groups (n=40) will be treated through structural diagnosis and management concept (SDM), regional approaches (RA), McKenzie mechanical diagnosis and therapy (MDT) concept prescribed by advanced practice physiotherapist (APP), or pain medications prescribed by the specialist physician for 4 weeks. The outcome will be evaluated in baseline, intermediate test (14 days), and post-treatment (28 days) through Brief Pain Inventory (BPI) for pain, Goniometer reading for cervical range of motion (ROM), and Neck disability index (NDI) as the primary outcome. The secondary outcome will be quality of life measured at baseline and post-treatment by the WHO quality of life questionnaire WHOQOL-BREF. Discussion: The study will compare the efficacy of the three physiotherapy approaches with pain medications when treating chronic cervical radiculopathy. The findings will provide evidence when demining the best conservative management approach for CR. Clinical Trial Registry India: CTRI/2022/03/040922 (08/03/2022)

show abstract

“…Convolutional neural networks (CNN) as the main model of deep learning along with other network models such as recurrent neural network (RNN) and long short-term memory network (LSTM) are extensively used in emotion recognition, providing many corresponding optimization models and good research results [11]. Reference [12] focused on the extraction of features with high differences in the recognition process and provided a strategy for feature coordination recognition. As a result, ResNet101 and training iteration increased the entropy and realized high-accuracy video performance recognition [13].…”

Section: Related Studiesmentioning

confidence: 99%

Multimodal modeling of human emotions using sound, image and text fusion

Hosseini

Yamaghani

Arabani

2023

Preprint

View full text Add to dashboard Cite

Multimodal emotion recognition and analysis is considered a developing research field. Improving the multimodal fusion mechanism plays a key role in the more detailed recognition of the recognized emotion. The present study aimed to optimize the performance of the emotion recognition system and presented a model for multimodal emotion recognition from audio, text, and video data. First, the data were fused as a combination of video and audio, then as a combination of audio and text as binary, and finally the results were fused together. The final output included audio, text, and video data taking common features into account. Then, the convolutional neural network, as well as long-term and short-term memory (CNN-LSTM), were used to extract audio. Next, the Inception-Res Net-v2 network was applied for extracting the facial expression in the video. The output fused data were utilized by LSTM as the input of the softmax classifier to recognize the emotion of audio and video features fusion. In addition, the CNN-LSTM was combined in the form of a binary channel for learning audio emotion features. Meanwhile, a Bi-LSTM network was used to extract the text features and softmax was used for classifying the fused features. Finally, the generated results were fused together for the final classification, and the logistic regression model was used for fusion and classification. The results indicated that the recognition accuracy of the proposed method in the IEMOCAP data set was 82.9.

show abstract

Batch Entropy Supervised Convolutional Neural Networks for Feature Extraction and Harmonizing for Action Recognition

Cited by 11 publications

References 68 publications

The Application of Multimedia and Deep Learning in the Integration of Professional and Innovative Education in Colleges

The Application of Multimedia and Deep Learning in the Integration of Professional and Innovative Education in Colleges

Physiotherapy versus pain medication for managing chronic cervical radiculopathy: protocol for a multi-arm parallel-group randomized clinical trial

Multimodal modeling of human emotions using sound, image and text fusion

Contact Info

Product

Resources

About