At present, the visual tracking method based on Transformer has made significant progress, and has shown excellent performance on various data sets. However, the traditional Transformer tracker pays too much attention...
详细信息
In the post-pandemic era, online courses have been adopted universally. Manually assessing online course teaching quality requires significant time and professional pedagogy experience. To address this problem, we des...
详细信息
ISBN:
(数字)9781665490627
ISBN:
(纸本)9781665490627
In the post-pandemic era, online courses have been adopted universally. Manually assessing online course teaching quality requires significant time and professional pedagogy experience. To address this problem, we design an evaluation protocol and propose a multimodal machinelearning framework1 for automated teaching quality assessment of one-to-many online instruction videos. Our framework evaluates online teaching quality from five aspects, namely Clarity, Classroom interaction, Technical management of online teaching, Empathy, and Time management. Our method includes mid-level behavior feature extraction, high-level interpretable feature extraction, and supervised learning prediction. Our automated multimodal teaching quality assessment system achieves comparable performance to human annotators on our one-to-many online instruction videos. For binary classification, the best average accuracy of five aspects is 0.898. For regression, the best average means square error is 0.527 on a 0-10 scale.
Human behaviour analysis is a difficult aspect to maintain for a normal human being. this human behaviour analysis is unpredictable generally. the alignment of advanced technology such as machinelearning is essential...
详细信息
In companies or any organizations, the data is collected in various formats along with government-approved identity documents such as Aadhaar, PAN, license, etc. in soft or hard copy format. they contain valuable info...
详细信息
Agriculture is considered one of the primary sectors in India. Especially rice is the primary stable food widely cultivated and consumed by 50% of the Indian population. Paddy crop is susceptible to several diseases l...
详细信息
Finger gesture recognition using surface electromyography (sEMG) became an efficient Human-Robot Interaction (HRI) solution. Although machinelearning (ML) techniques are widely applied in this field, the general solu...
详细信息
ISBN:
(纸本)9781665486415
Finger gesture recognition using surface electromyography (sEMG) became an efficient Human-Robot Interaction (HRI) solution. Although machinelearning (ML) techniques are widely applied in this field, the general solutions for labeling and collecting big datasets impose time-consuming implementation and heavy workloads. In this paper, a new deep learning structure, namely three-dimensional convolutional long short-term memory neural networks (3D-CLDNN) for finger gesture identification based on depth vision and sEMG signals, was proposed for human-machine interaction. It automatically labels the depthdata by the self-organizing map (SOM) and predicts the hand gesture only adopting sEMG signals. the 3D-CLDNN method is integrated to improve the recognition rate and computational speed. the results showed the highest clustering accuracy (98.60%) and highest accuracy (84.40%) withthe lowest computational time compared with different approaches. Finally, real-time human-machine interaction experiments are performed to demonstrate its efficiency.
Withthe development of society, the traditional retail industry has evolved into a pattern of coexistence of chain supermarkets, e-commerce sales, and unmanned retail, which greatly facilitates people's lives. Ho...
详细信息
Feature selection is critical in fields like datamining and pattern classification, as it eliminates irrelevant data and enhances the quality of highly dimensional datasets. this study explores the effectiveness of t...
详细信息
Feature selection is critical in fields like datamining and pattern classification, as it eliminates irrelevant data and enhances the quality of highly dimensional datasets. this study explores the effectiveness of the Q-learning embedded sine cosine algorithm (QLESCA) for feature selection in industrial casting defect detection using the VGG19 model. QLESCA's performance is compared to other optimization algorithms, with experimental results showing that QLESCA outperforms the other algorithms in terms of classification metrics. the best accuracy achieved by QLESCA is 97.0359%, with an average fitness value of - 0.99124. the proposed method provides a promising approach to improve the accuracy and reliability of industrial casting defect detection systems, which is essential for product quality and safety. Our findings suggest that using powerful optimization algorithms like QLESCA is crucial for obtaining the best subsets of information in feature selection and achieving optimal performance in classification tasks.
Breast Cancer disease is the utmost characterized heterogeneous illnesses consisting of various types. Apart from lung cancer, Breast cancer is spreading widely everywhere. this research work confines to accurately an...
详细信息
the reliability assessment of a machinelearning model's prediction is an important quantity for the deployment in safety critical applications. Not only can it be used to detect novel sceneries, either as out-of-...
详细信息
ISBN:
(数字)9781665490627
ISBN:
(纸本)9781665490627
the reliability assessment of a machinelearning model's prediction is an important quantity for the deployment in safety critical applications. Not only can it be used to detect novel sceneries, either as out-of-distribution or anomaly sample, but it also helps to determine deficiencies in the training data distribution. A lot of promising research directions have either proposed traditional methods like Gaussian processes or extended deep learning based approaches, for example, by interpreting them from a Bayesian point of view. In this work we propose a novel approach for uncertainty estimation based on autoencoder models: the recursive application of a previously trained autoencoder model can be interpreted as a dynamical system storing training examples as attractors. While input images close to known samples will converge to the same or similar attractor, input samples containing unknown features are unstable and converge to different training samples by potentially removing or changing characteristic features. the use of dropout during training and inference leads to a family of similar dynamical systems, each one being robust on samples close to the training distribution but unstable on new features. Either the model reliably removes these features or the resulting instability can be exploited to detect problematic input samples. We evaluate our approach on several dataset combinations as well as on an industrial application for occupant classification in the vehicle interior for which we additionally release a new synthetic dataset.
暂无评论