scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Dancer Tracking Algorithm in Ethnic Areas Based on Multifeature Fusion Neural Network

TL;DR: The tracking algorithm proposed in this paper has higher robustness than other algorithms and effectively reduces the error samples generated during the tracking process, thus improving the accuracy of long-term tracking.
Abstract: Due to the complex posture changes in dance movements, accurate detection and tracking of human targets are carried out in order to improve the guidance ability of dancers in ethnic areas. A multifeature fusion-based tracking algorithm for dancers in ethnic areas is proposed. The edge contour model of video images of dancers in ethnic areas is detected, and the video tracking scanning imaging model of dancers in ethnic areas is constructed. The video images of dancers in ethnic areas are enhanced based on the initial contour distribution, and a visual perception model of dancers tracking images in ethnic areas is established. To improve the algorithm’s estimation of complex poses and finally complete the dance movement recognition, a feature pyramid network is used to extract the features of dance movements, and then, a multifeature fusion module is used to fuse multiple features. The tracking algorithm proposed in this paper has higher robustness than other algorithms and effectively reduces the error samples generated during the tracking process, thus improving the accuracy of long-term tracking.
Citations
More filters
Journal ArticleDOI
TL;DR: In this paper , the authors proposed a method to solve the problem of the problem: the one-dimensional graph. .>

Abstract:
References
More filters
Journal ArticleDOI
TL;DR: A human-related multi-stream CNN (HR-MSCNN) architecture that encodes appearance, motion, and the captured tubes of the human- related regions is introduced that achieves state-of-the-art results on these four datasets.

182 citations

Journal ArticleDOI
TL;DR: This study constructs a mapping between visual features and a semantic descriptor of each action category, allowing new categories to be recognised in the absence of any visual training data, and achieves the state-of-the-art zero-shot action recognition performance with a simple and efficient pipeline, and without supervised annotation of attributes.
Abstract: The number of categories for action recognition is growing rapidly and it has become increasingly hard to label sufficient training data for learning conventional models for all categories. Instead of collecting ever more data and labelling them exhaustively for all categories, an attractive alternative approach is "zero-shot learning" (ZSL). To that end, in this study we construct a mapping between visual features and a semantic descriptor of each action category, allowing new categories to be recognised in the absence of any visual training data. Existing ZSL studies focus primarily on still images, and attribute-based semantic representations. In this work, we explore word-vectors as the shared semantic space to embed videos and category labels for ZSL action recognition. This is a more challenging problem than existing ZSL of still images and/or attributes, because the mapping between video space-time features of actions and the semantic space is more complex and harder to learn for the purpose of generalising over any cross-category domain shift. To solve this generalisation problem in ZSL action recognition, we investigate a series of synergistic strategies to improve upon the standard ZSL pipeline. Most of these strategies are transductive in nature which means access to testing data in the training phase. First, we enhance significantly the semantic space mapping by proposing manifold-regularized regression and data augmentation strategies. Second, we evaluate two existing post processing strategies (transductive self-training and hubness correction), and show that they are complementary. We evaluate extensively our model on a wide range of human action datasets including HMDB51, UCF101, Olympic Sports and event datasets including CCV and TRECVID MED 13. The results demonstrate that our approach achieves the state-of-the-art zero-shot action recognition performance with a simple and efficient pipeline, and without supervised annotation of attributes. Finally, we present in-depth analysis into why and when zero-shot works, including demonstrating the ability to predict cross-category transferability in advance.

131 citations

Journal ArticleDOI
TL;DR: Time Difference of Arrival (TDoA) localization accuracy, update probability, and update frequency were evaluated for different trajectories (walking, cycling, and driving) and LoRa spreading factors.
Abstract: The performance of LoRa geolocation for outdoor tracking purposes has been investigated on a public LoRaWAN network. Time Difference of Arrival (TDoA) localization accuracy, update probability, and update frequency were evaluated for different trajectories (walking, cycling, and driving) and LoRa spreading factors. A median accuracy of 200 m was obtained for the raw TDoA output data. In 90% of the cases, the error was less than 480 m. Taking into account the road map and movement speed significantly improves accuracy to a median of 75 m and a 90th percentile error of less than 180 m.

97 citations

Journal ArticleDOI
TL;DR: A fine-grained visual recognition model named as MCF-Net to classifying different crop species in practical farmland scenes is presented, acceptable and suitable to the implementation of IoT platforms in precision agricultural practices.

90 citations

Journal ArticleDOI
27 Feb 2022-Agronomy
TL;DR: A Reversible Automatic Selection Normalization (RASN) network is proposed, integrating the normalization and renormalization layer to evaluate and select thenormalization module of the prediction model, showing good prediction ability and adaptability for the greenhouse in the Smart Agriculture System.
Abstract: Due to the nonlinear modeling capabilities, deep learning prediction networks have become widely used for smart agriculture. Because the sensing data has noise and complex nonlinearity, it is still an open topic to improve its performance. This paper proposes a Reversible Automatic Selection Normalization (RASN) network, integrating the normalization and renormalization layer to evaluate and select the normalization module of the prediction model. The prediction accuracy has been improved effectively by scaling and translating the input with learnable parameters. The application results of the prediction show that the model has good prediction ability and adaptability for the greenhouse in the Smart Agriculture System.

70 citations