The traditional color system of Beijing Forbidden City is comprehensive and complex. The research focuses on collecting color data from the Forbidden City, analyzing subjective and objective color attributes, and comp...
详细信息
ISBN:
(数字)9798350380347
ISBN:
(纸本)9798350380354
The traditional color system of Beijing Forbidden City is comprehensive and complex. The research focuses on collecting color data from the Forbidden City, analyzing subjective and objective color attributes, and compute architectural color image scale. First this research used K-means algorithm to extract the theme color cards of Beijing Forbidden City. Subsequently, single colors were extracted using the idea of community detection, and matching colors were extracted based on the thought of association rule mining. Then this research extracted objective attributes of colors and designed experiments to quantify subjective attributes. Finally, the color subjective and objective data fitting was completed. This outcome is helpful for the digital preservation of traditional colors and the advancement of color culture inheritance practices.
The creation of Chinese dance dramas plays an important role in the dissemination of Chinese culture in the context of pluralistic unity. As one of the outstanding poetry dramas in the past two years, “The Journey of...
详细信息
ISBN:
(数字)9798350380347
ISBN:
(纸本)9798350380354
The creation of Chinese dance dramas plays an important role in the dissemination of Chinese culture in the context of pluralistic unity. As one of the outstanding poetry dramas in the past two years, “The Journey of a Legendary Landscape Painting” (JLLP) has rich national cultural connotations and aesthetic value in its stage costume colors. This article first studied the data extraction method of stage costume colors based on clustering and association rule mining, and constructed the color dataset for JLLP. On this basis, the color semantic mining and extraction including subjective and objective factors were conducted. Finally, combined with color semantics, the character personality of JLLP was further analyzed. The research results show that the color extraction method designed in this article can accurately preserve the color characteristics of stage costumes. In addition, the costume colors of JLLP play a good supporting role in shaping the character personality.
Oral history involves conveying experienced events through spoken narratives, typically preserved and later studied by converting spoken language into text. Named entity recognition (NER) is the recognition of meaning...
详细信息
ISBN:
(数字)9798350380347
ISBN:
(纸本)9798350380354
Oral history involves conveying experienced events through spoken narratives, typically preserved and later studied by converting spoken language into text. Named entity recognition (NER) is the recognition of meaningful named entities in text, such as person names, locations, organizations, etc. Conducting NER research on oral history text is significant as it not only facilitates the verification of named entities by researchers but also allows for the extraction of key information from texts through named entities. By collecting oral history corpus, we establish an oral history text dataset, manually annotating all texts with six entity labels and repeatedly verifying them manually. Utilizing the concept of machine reading comprehension (MRC), a semantic fusion module was added, integrating label information into text information using a cross-attention mechanism, and decoding the corresponding entity labels through span decoding. The final model performs well on the oral history text dataset and also showed good performance on general Chinese datasets.
Music recommendation algorithms aim to predict the songs that users might be interested in based on various factors, such as their listening history. The best existing method is Bert_Music, a sequential music recommen...
Music recommendation algorithms aim to predict the songs that users might be interested in based on various factors, such as their listening history. The best existing method is Bert_Music, a sequential music recommendation model based on Bidirectional Encoder Representation from Transformer (BERT). This model leverages user sequence context information to improve recommendation performance. However, Bert_Music only considers the sequence information of songs and ignores their content characteristics. To address this limitation, we propose an improved version of Bert_Music, which is a sequential music recommendation model based on Bidirectional Encoder Representation from Transformer with tag information (BERT_MUSIC_T), uses song ID (Identity) and tag information to make more accurate recommendations. We compared the performance of Bert_Music_T with that of the Bert_Music model on the *** dataset and found that the Precision has increased by up to 2.85%. These results prove the validity of our method.
In order to solve the problem of fast switching of special equipment in the cultural complex, the improved Moth-Flame Optimization algorithm (MFO) is combined with the improved Dynamic Window Approach (DWA) to realize...
In order to solve the problem of fast switching of special equipment in the cultural complex, the improved Moth-Flame Optimization algorithm (MFO) is combined with the improved Dynamic Window Approach (DWA) to realize the hybrid path planning of handling robots in different scenarios of cultural complexes in this paper. Specifically, the traditional Moth-Flame Optimization algorithm is improved and applied to the global path planning problem. On the basis of the planned global path, the intersection of dynamic obstacles with the global path is detected, and the improved Dynamic Window Approach is used for local obstacle avoidance in the section with collision risk, finally, the hybrid path planning in the complex environment of the cultural complex is realized. The simulation experimental results prove the effectiveness of the proposed algorithm.
Music source separation, the process of extracting independent audio streams from a complex mix, has traditionally focused on isolating vocals, drums, bass, and other primary sources. This study tackles the more intri...
详细信息
ISBN:
(数字)9798350380347
ISBN:
(纸本)9798350380354
Music source separation, the process of extracting independent audio streams from a complex mix, has traditionally focused on isolating vocals, drums, bass, and other primary sources. This study tackles the more intricate task of separating the piano component from a piano concerto-a challenge compounded by the diverse range of instruments and the dynamic shifts in volume and timbre. Unlike traditional music separation tasks, the piano’s distinct characteristics and its interaction with the orchestra demand a more nuanced *** address the scarcity of multi-track recordings for piano concertos, this research pioneers an artificial data synthesis strategy to create a robust training dataset. We introduce a novel hybrid deep learning model that integrates Long Short-Term Memory (LSTM) networks with Transformer architecture, capitalizing on their complementary strengths to distinguish piano melodies from the rich tapestry of orchestral sounds. Our experiments demonstrate that this hybrid approach significantly outperforms conventional methods, with an improvement of 3.18 dB in signal-to-distortion ratio. These results not only validate the efficacy of proposed method but also pave the way for innovative applications in classical music source separation.
Dunhuang mural images classification belongs to the research task in the field of image recognition. In this paper, the semi-supervised model is established with multidimensional features extracted by transfer learnin...
详细信息
Dunhuang mural images classification belongs to the research task in the field of image recognition. In this paper, the semi-supervised model is established with multidimensional features extracted by transfer learning. A small number of labeled samples were used to obtain a large number of unlabeled data, combined with Active Learning and iterative strategy for multiple rounds of label transfer of selected samples. After several rounds of iterations, we can get a more powerful classification learner. Experiments on the self-built Dunhuang mural dataset have shown that the results can approach or even exceed some supervised learning methods when the number of known label samples is less than 4%. Our research implements the cultural resource image classification algorithm based on small samples, which is conductive to improving the accuracy when label samples are scarce.
Beatbox is a performance form of vocal percussion, which is gradually favored by people with continuous development in recent years. With the abundance of Beatbox content on the Internet, the research on its automatic...
详细信息
ISBN:
(数字)9781728181387
ISBN:
(纸本)9781728181394
Beatbox is a performance form of vocal percussion, which is gradually favored by people with continuous development in recent years. With the abundance of Beatbox content on the Internet, the research on its automatic classification becomes very significant. In this article, we mainly using machine learning methods, to identify the beatbox timbre. First of all, we collected 900 Beatbox audio materials and constructed a database. Secondly, 14 acoustic parameters of audio material are extracted by the signal processing method. Finally, we used MultilayerPerceptron, SMO, BayesNet, RandomForest in current machine learning to build a training model for BB timbre classification. The Experimental results showed that the model could classify BB timbre well.
In view of the problems of low equipment integration, difficult equipment operation, and high technical threshold existing in the current stage visual effect linkage controlsystem, a highly integrated linkage control...
详细信息
ISBN:
(数字)9781728181387
ISBN:
(纸本)9781728181394
In view of the problems of low equipment integration, difficult equipment operation, and high technical threshold existing in the current stage visual effect linkage controlsystem, a highly integrated linkage controlsystem for stage visual effect equipment, which is suitable for small and medium-sized performance venues, is proposed in this paper. This system can control the stage lighting and the playback of multi-channel and multi-format media content, and it is easy to control. The design scheme of the system is described in detail, and the feasibility of the system is verified through tests in this paper.
Some exiting works have explored the promotion between depth estimation and semantic segmentation. These works are usually based on convolutional neural networks, which extract compact features and map the relationshi...
详细信息
ISBN:
(数字)9781728181387
ISBN:
(纸本)9781728181394
Some exiting works have explored the promotion between depth estimation and semantic segmentation. These works are usually based on convolutional neural networks, which extract compact features and map the relationship between input and output according to specific tasks. In this paper, we introduce a novel adversarial training strategy, that is, generator produces the semantic segmentation map and depth map, and then the discriminator can judge the authenticity of the synthesized color image, thereby supervising the output of the network during back propagation. We use the adversarial loss combined with the reconstruction loss function to supervise the model, and find that the adversarial loss function which is seen as a global supervision can further optimize the output. We use 200 color images from Kitti dataset with semantic segmentation ground truth as the training set, and train the network in an end-to-end manner. The experimental results show that the adversarial training method is well applied to the multi-task training combining semantic segmentation and depth estimation, and can further improve the quantitative performance of depth estimation.
暂无评论