文献类型： 会议论文
第一作者： Hajer Essefi
作者： Hajer Essefi 1 ; Olfa Ben Ahmed 1 ; Christel Bidet-Ildei 2 ; Yannick Blandin 2 ; Christine Fernandez-Maloigne 1 ;
作者机构： 1.XLIM Research Institute, UMR CNRS 7252, University of Poitiers
2.Centre de Recherches sur la Cognition et l'Apprentissage (UMR CNRS 7295), Universite de Poitiers, Universite de Tours, Centre National de la Recherche Scientifique
关键词： Deep learning;Computer vision;Human action recognition;Action perception;RGB video
会议名称： International Conference on Agents and Artificial Intelligence
主办单位：
页码： 351-359
摘要： Human Action Recognition (HAR) is an important task for numerous computer vision applications. Recently, deep learning approaches have shown proficiency in recognizing actions in RGB video. However, existing models rely mainly on global appearance and could potentially under perform in real world applications, such as sport events and clinical applications. Refereeing to domain knowledge in how human perceive action, we hypothesis that observing the dynamic of a 2D human body joints representation extracted from RGB video frames is sufficient to recognize an action in video. Moreover, body joints contain structural information with a strong spatial (intra-frame) and temporal (inter-frame) correlation between adjacent joints. In this paper, we propose a psychology-inspired twin stream Gated Recurrent Unit network for action recognition based on the dynamic of 2D human body joints in RGB videos. The proposed model achieves a classification accuracy of 89, 97% in a subject-specific experiment and outperforms the baseline method that fuses depth and inertial sensor data on the UTD-MHAD dataset. The proposed framework is more cost effective and highly competitive than depth 3D skeleton based solutions and therefore can be used outside capture motion labs for real world applications.
分类号： TP18-53
- 相关文献
[1]A Computer Vision Approach to Radial Velocity Extraction for Exoplanet Detection. Katelyn Gan,Vinesh Maguire Rajpaul. 2023
[2]Study for Emotion Recognition of Different Age Groups Students during Online Class. Ati Jain,Hare Ram Sah,Abhay Kothari. 2021
[3]Automated Tomato Inspection and Harvesting System Using Robotic Arm and Computer Vision in Greenhouses. Gaby Abou Haidar,Maher Alaa Deen,Roger Achkar,Michel Owayjan,Roy Abi Zeid Daou. 2023
[4]Detecting Vehicular Networking Node Misbehaviour Using Machine Learning. Saleha Saudagar,Rekha Ranawat. 2023
[5]Deep Learning Model Integrating Dilated Convolution and Deep Supervision for Brain Tumor Segmentation in Multi-parametric MRI. Tongxue Zhou,Su Ruan,Haigen Hu,Stephane Canu. 2019
[6]Approaches for Identifying Suicide Ideation in Social Media Texts: Comprehensive Review. Jayshri Suresh Sonawane,Dinesh Jain. 2024
[7]Pneumonia Detection and Chest X-Rays: Comprehensive Analysis of Artificial Intelligence Techniques in Clinical and Radiological Insights. Mohini Gahlot,Pinaki Ghosh. 2024
[8]Detection of Emotions from Speech using Deep Learning Techniques and Traditional Techniques: A Survey. Rashmi Rani,Manoj Kumar Ramaiya. 2023
[9]A Comparative Study of Various Learning Models for Object Detection in Contextual Scene Interpretation. Taranpreet Singh,Dr.Hemang Shrivastava. 2023
[10]APPLICATION OF U-NET CONVOLUTIONAL NEURAL NETWORK TO BUSHFIRE MONITORING IN AUSTRALIA WITH SENTINEL-1/-2 DATA. Isabella. K. Lee,John. C. Trinder,Arcot. Sowmya. 2020
[11]Student’s Feedback by emotion and speech recognition through Deep Learning. Ati Jain,Hare Ram Sah. 2021
[12]A Review of Protein Sequences of COVID-19 Using Machine Learning and Deep Learning Approaches. Anurag Golwelkar,Abhay Kothari. 2023
