Abstract This paper explores a Sports movement trajectory data capturing technique based on the mean shift algorithm to overcome the challenges of collecting information in complex environmental conditions and high speed movements. The method constructs a framework with 51 degrees of freedom and 16 joints to collect and analyze trajectory information during training and performs corresponding dimensionality reduction for better results.... This paper explores a Sports movement trajectory data capturing technique based on the mean shift algorithm to overcome the challenges of collecting information in complex environmental conditions and high speed movements. The method constructs a framework with 51 degrees of freedom and 16 joints to collect and analyze trajectory information during training and performs corresponding dimensionality reduction for better results. To reduce the dependency of the mean shift algorithm on environmental parameters, the probability density function in the gradient iteration estimation algorithm is chosen, and the color information of the target is used as a feature for trajectory data capturing. Experimental results demonstrate that this method effectively captures the joint activities of athletes, making Sports movement training trajectory data more accurate without any external parameter intervention. Read More Read Less

Abstract Educational Quality assessment serves as a crucial instrument for enhancing teaching quality and bolstering teaching effectiveness. Conventional college English teaching evaluations are static assessments that fail to reflect student's performance throughout the teaching process accurately. Hence, this paper develops a college English teaching evaluation model founded on deep learning neural networks. It achieves emotion classification via facial recognition of students... Educational Quality assessment serves as a crucial instrument for enhancing teaching quality and bolstering teaching effectiveness. Conventional college English teaching evaluations are static assessments that fail to reflect student's performance throughout the teaching process accurately. Hence, this paper develops a college English teaching evaluation model founded on deep learning neural networks. It achieves emotion classification via facial recognition of students and integrates this with a standard distribution evaluation model to assess student's attitudes toward English teaching quality. The experimental outcomes reveal that the proposed model significantly enhances the accuracy of emotion recognition and classification rates, effectively mirroring student's attitudes towards English instruction in real world applications. Read More Read Less

Abstract This paper proposes a novel multimodal music emotion classification algorithm that integrates audio and lyrical features to overcome the limitations of single modality approaches. Recognizing that music conveys emotion through both sound and text, the system employs deep learning techniques, specifically combining One D CNN and TwoD CNN models with C3D and I3D frameworks for audio processing, alongside text analysis... This paper proposes a novel multimodal music emotion classification algorithm that integrates audio and lyrical features to overcome the limitations of single modality approaches. Recognizing that music conveys emotion through both sound and text, the system employs deep learning techniques, specifically combining One D CNN and TwoD CNN models with C3D and I3D frameworks for audio processing, alongside text analysis using TF IDF and Word2vec. To effectively fuse these heterogeneous modalities, the study implements a stacking based decision level fusion strategy with a Soft Max secondary classifier, significantly outperforming featurelevel and traditional decision fusion methods. Utilizing transfer learning on datasets like Sport-1M and Kinetics enhances model generalization, while Adam and SGD optimizers improve training efficiency. Experimental results on a dataset of 2000 songs (anger, happiness, relaxation, sadness) demonstrate that the proposed multimodal approach achieves a maximum classification accuracy of 78%, a 4% improvement over singlemodal classifiers and a 2% gain over other fusion techniques. The method effectively mitigates data heterogeneity and over fitting via 5-fold cross validation and addresses challenges in classifying “relaxation” by leveraging complementary audio lyric cues. Evaluation metrics (accuracy, F1-score, ROC/AUC) confirm superior performance, validating that synergistic integration of audio spectral features and semantic lyric representations yields more precise, robust, and scalable music emotion recognition, even with limited labeled data. Read More Read Less