Person re-identification aims to retrieve pedestrians with the same identity across different cameras. However, current methods increase attention to interfering regions when dealing with complex backgrounds and occlusion, especially in the presence of similar interfering features. To enhance the robustness of the model, we propose the Reverse Pyramid Attention Guidance (RPAG) network, using a reverse pyramid structure to learn features at multiple granularities. To mitigate the impact of occlusion, we introduce the Similar Feature Filtering (SFF) attention module at the pixel level, using graph convolution to adaptively select occluded regions, thereby enhancing retrieval accuracy by filtering out irrelevant parts. Combining the reverse pyramid structure with the pixel-level attention module strengthens adaptability to complex scenes, guides multi-granularity feature learning, and effectively handles various occlusion scenarios. RPAG achieved Rank-1 accuracies of 96.2%, 93.2%, 88.7%, and 73.2% on the Market1501, DukeMTMC-ReID, MSMT17, and Occluded-Duke datasets, respectively.
{"title":"Reverse Pyramid Attention Guidance Network for Person Re-Identification","authors":"Jiang Liu, Wei Bai, Yun Hui","doi":"10.4018/ijcini.349982","DOIUrl":"https://doi.org/10.4018/ijcini.349982","url":null,"abstract":"Person re-identification aims to retrieve pedestrians with the same identity across different cameras. However, current methods increase attention to interfering regions when dealing with complex backgrounds and occlusion, especially in the presence of similar interfering features. To enhance the robustness of the model, we propose the Reverse Pyramid Attention Guidance (RPAG) network, using a reverse pyramid structure to learn features at multiple granularities. To mitigate the impact of occlusion, we introduce the Similar Feature Filtering (SFF) attention module at the pixel level, using graph convolution to adaptively select occluded regions, thereby enhancing retrieval accuracy by filtering out irrelevant parts. Combining the reverse pyramid structure with the pixel-level attention module strengthens adaptability to complex scenes, guides multi-granularity feature learning, and effectively handles various occlusion scenarios. RPAG achieved Rank-1 accuracies of 96.2%, 93.2%, 88.7%, and 73.2% on the Market1501, DukeMTMC-ReID, MSMT17, and Occluded-Duke datasets, respectively.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141924145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The relationship between the emotional components associated with images and text is a crucial way of multimodal emotion analysis. However, most of the present multimodel affective cognitive models simply associate the features of images and texts without thoroughly investigating their interactions, resulting in poor recognition. Therefore, a multimodel emotion cognition method based on multi-channel graphic interaction is proposed. Text context features are extracted, scene and image information is encoded, and useful features are obtained. Based on these results, the modal alignment module be applied to obtain information about affective regions and words, and then the cross-modal gating module be applied to combine the multimodel features. In addition, we tested extensively on three open datasets, achieving an accuracy of 0.8122 for the MSA-single dataset, 0.7307 for the MSA-MULTIPLE dataset, and 0.7159 for TumEmo. The results show that this method is effective for multimodal emotion detection.
{"title":"Multimodal Emotion Cognition Method Based on Multi-Channel Graphic Interaction","authors":"Baisheng Zhong","doi":"10.4018/ijcini.349969","DOIUrl":"https://doi.org/10.4018/ijcini.349969","url":null,"abstract":"The relationship between the emotional components associated with images and text is a crucial way of multimodal emotion analysis. However, most of the present multimodel affective cognitive models simply associate the features of images and texts without thoroughly investigating their interactions, resulting in poor recognition. Therefore, a multimodel emotion cognition method based on multi-channel graphic interaction is proposed. Text context features are extracted, scene and image information is encoded, and useful features are obtained. Based on these results, the modal alignment module be applied to obtain information about affective regions and words, and then the cross-modal gating module be applied to combine the multimodel features. In addition, we tested extensively on three open datasets, achieving an accuracy of 0.8122 for the MSA-single dataset, 0.7307 for the MSA-MULTIPLE dataset, and 0.7159 for TumEmo. The results show that this method is effective for multimodal emotion detection.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141922721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Feature selection aims to select a subset of features from high-dimensional data, which can overcome the curse of dimensionality for the next dealing steps. However, the feature selection itself could face the curse of dimensionality. To overcome the above problem, in this paper, a new feature selection framework is designed according to a human processing in our daily life. In our daily life, to evaluate a candidate's ability to work, the related professional knowledge and the comprehensive ability of a candidate should be both evaluated. Actually, a candidate only with good professional knowledge often hardly solves new problems in the work. Based on the above analysis, in our new designed framework, the features are selected by evaluating its ability of global structure preservation and self-representation, which are respectively similar to the professional knowledge and comprehensive ability in evaluating candidate. As a result, the selected features can accommodate larger changes in test data. The conducted experiments validate the effectiveness of our feature selection.
{"title":"Global Structure Preservation and Self-Representation-Based Supervised Feature Selection","authors":"Qing Ye, Yaxin Sun","doi":"10.4018/ijcini.346987","DOIUrl":"https://doi.org/10.4018/ijcini.346987","url":null,"abstract":"Feature selection aims to select a subset of features from high-dimensional data, which can overcome the curse of dimensionality for the next dealing steps. However, the feature selection itself could face the curse of dimensionality. To overcome the above problem, in this paper, a new feature selection framework is designed according to a human processing in our daily life. In our daily life, to evaluate a candidate's ability to work, the related professional knowledge and the comprehensive ability of a candidate should be both evaluated. Actually, a candidate only with good professional knowledge often hardly solves new problems in the work. Based on the above analysis, in our new designed framework, the features are selected by evaluating its ability of global structure preservation and self-representation, which are respectively similar to the professional knowledge and comprehensive ability in evaluating candidate. As a result, the selected features can accommodate larger changes in test data. The conducted experiments validate the effectiveness of our feature selection.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141929342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
At present, the combination of general evolutionary algorithms (EAs) and neural networks is limited to optimizing the framework or hyper parameters of neural networks. To further extend applications of EAs on neural networks, we propose a particle swarm optimization (PSO) based generative adversarial network(GAN), named as PGAN in this paper. In the study, PSO is utilized as a generator to generate fake data, while the discriminator is a traditional fully connected neural network. In the confrontation process, when the proposed PSO can generate a better fake image, this will react to the discriminator, so that the discriminator can improve the recognition effect of the image and the better discriminator also accelerates the evolution of the overall model. Through experiments, we explore the new application value of EAs in deep learning, so that the sample data in EAs and the sample data in deep learning are interconnected. The PSO algorithm is improved, so that it truly participates in the confrontation with multi-layer perceptrons.
目前，一般进化算法（EA）与神经网络的结合仅限于优化神经网络的框架或超参数。为了进一步扩展进化算法在神经网络中的应用，我们提出了一种基于粒子群优化（PSO）的生成式对抗网络（GAN），本文将其命名为 PGAN。在研究中，PSO 被用作生成器来生成虚假数据，而判别器则是传统的全连接神经网络。在对抗过程中，当所提出的 PSO 能够生成较好的假图像时，就会对判别器产生反应，从而使判别器提高图像的识别效果，而较好的判别器也会加速整个模型的演化。通过实验，我们探索了 EAs 在深度学习中新的应用价值，使 EAs 中的样本数据与深度学习中的样本数据相互关联。改进 PSO 算法，使其真正参与到与多层感知器的对抗中。
{"title":"A Particle Swarm Optimization-Based Generative Adversarial Network","authors":"Haojie Song, Xuewen Xia, Lei Tong","doi":"10.4018/ijcini.349935","DOIUrl":"https://doi.org/10.4018/ijcini.349935","url":null,"abstract":"At present, the combination of general evolutionary algorithms (EAs) and neural networks is limited to optimizing the framework or hyper parameters of neural networks. To further extend applications of EAs on neural networks, we propose a particle swarm optimization (PSO) based generative adversarial network(GAN), named as PGAN in this paper. In the study, PSO is utilized as a generator to generate fake data, while the discriminator is a traditional fully connected neural network. In the confrontation process, when the proposed PSO can generate a better fake image, this will react to the discriminator, so that the discriminator can improve the recognition effect of the image and the better discriminator also accelerates the evolution of the overall model. Through experiments, we explore the new application value of EAs in deep learning, so that the sample data in EAs and the sample data in deep learning are interconnected. The PSO algorithm is improved, so that it truly participates in the confrontation with multi-layer perceptrons.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141799711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Traditional evaluation methods based on static elastic-plastic analysis usually produce results with low fitting degree compared with actual data. In this paper, a new seismic performance evaluation method is proposed. Firstly, several evaluation indexes of green ecological building structure are calculated. Subsequently, a cumulative damage model is established based on fatigue life curve and Miner criterion, which allows to determine the total dissipated strain energy and damage index. In order to verify the effectiveness of this method, it is compared with the traditional static elastic-plastic analysis method through experimental tests. The results show that compared with the traditional method of 50% to 60%, the proposed method achieves a significantly higher fitting degree with the actual data, ranging from 70% to 92%. This emphasizes the superiority and reliability of the proposed method in evaluating the seismic performance of green ecological building structures, and provides insights for safer and more flexible building practices.
{"title":"Study on Multi-Index Evaluation Technology of Seismic Performance of Green Ecological Building Structure","authors":"Lei Zhang, Xin-Yi Huang, Hui Sun","doi":"10.4018/ijcini.349936","DOIUrl":"https://doi.org/10.4018/ijcini.349936","url":null,"abstract":"Traditional evaluation methods based on static elastic-plastic analysis usually produce results with low fitting degree compared with actual data. In this paper, a new seismic performance evaluation method is proposed. Firstly, several evaluation indexes of green ecological building structure are calculated. Subsequently, a cumulative damage model is established based on fatigue life curve and Miner criterion, which allows to determine the total dissipated strain energy and damage index. In order to verify the effectiveness of this method, it is compared with the traditional static elastic-plastic analysis method through experimental tests. The results show that compared with the traditional method of 50% to 60%, the proposed method achieves a significantly higher fitting degree with the actual data, ranging from 70% to 92%. This emphasizes the superiority and reliability of the proposed method in evaluating the seismic performance of green ecological building structures, and provides insights for safer and more flexible building practices.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141800112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the continuous development of society and economy, Internet thinking is gradually integrated into interior design, bringing new possibilities for interior design. This paper takes the interior design of general hospital outpatient department based on Internet thinking as an example to explore the interior design based on Internet thinking. Through the questionnaire survey and interview method and other scientific research methods, the author collected relevant data and carried out statistics and analyses, and concluded that the field of interior design should be combined with the current way of thinking about the Internet to innovate, and integrate the interior space into the combination of modern science and technology and interior design, so as to enable the interior design industry to develop rapidly. Under the influence of Internet thinking, interior design pays more attention to the interactive relationship between users and space. This mode of thinking enables designers to better understand user needs and provide users with more personalised and comfortable space design.
{"title":"Interior Design Based on Internet Thinking","authors":"Min Nie","doi":"10.4018/ijcini.349971","DOIUrl":"https://doi.org/10.4018/ijcini.349971","url":null,"abstract":"With the continuous development of society and economy, Internet thinking is gradually integrated into interior design, bringing new possibilities for interior design. This paper takes the interior design of general hospital outpatient department based on Internet thinking as an example to explore the interior design based on Internet thinking. Through the questionnaire survey and interview method and other scientific research methods, the author collected relevant data and carried out statistics and analyses, and concluded that the field of interior design should be combined with the current way of thinking about the Internet to innovate, and integrate the interior space into the combination of modern science and technology and interior design, so as to enable the interior design industry to develop rapidly. Under the influence of Internet thinking, interior design pays more attention to the interactive relationship between users and space. This mode of thinking enables designers to better understand user needs and provide users with more personalised and comfortable space design.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141802236","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As China develops and grows, there are an increasing number of Chinese language learners worldwide, and Chinese learners are becoming more and more interested in the Chinese Proficiency Test (HSK). The Chinese Proficiency Test (HSK) is also continuously evolving and improving. After more than 20 years of development, it has gradually been recognized by Chinese language learners at home and abroad. However, due to the short launch time, the existence of deficiencies is inevitable. But the IELTS exam was introduced in China in 1987 and has since amassed a wealth of knowledge and practice. Its framework is similar to HSK, so it can be used for HSK propositions and tests. One of the most challenging aspects of teaching Chinese is writing which involves the writing of Chinese characters, most Chinese learners have difficulty in writing. This paper mainly compares the test design and content in the writing part between HSK and IELTS, analyzes the effectiveness of the writing part of HSK, and puts forward some reference suggestions for the future writing test proposition and implementation of HSK.
{"title":"Effectiveness Analysis of IELTS and HSK Test Based on SVM in the Development of English Writing Ability","authors":"Yan Li","doi":"10.4018/ijcini.346227","DOIUrl":"https://doi.org/10.4018/ijcini.346227","url":null,"abstract":"As China develops and grows, there are an increasing number of Chinese language learners worldwide, and Chinese learners are becoming more and more interested in the Chinese Proficiency Test (HSK). The Chinese Proficiency Test (HSK) is also continuously evolving and improving. After more than 20 years of development, it has gradually been recognized by Chinese language learners at home and abroad. However, due to the short launch time, the existence of deficiencies is inevitable. But the IELTS exam was introduced in China in 1987 and has since amassed a wealth of knowledge and practice. Its framework is similar to HSK, so it can be used for HSK propositions and tests. One of the most challenging aspects of teaching Chinese is writing which involves the writing of Chinese characters, most Chinese learners have difficulty in writing. This paper mainly compares the test design and content in the writing part between HSK and IELTS, analyzes the effectiveness of the writing part of HSK, and puts forward some reference suggestions for the future writing test proposition and implementation of HSK.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141831043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xuezhi Yue, Ye Fan, Yuan Zeng, Weitao Fan, Luhui Zhou
This article analyzes the commonly used collision detection methods in game engines and designs a hybrid bounding box structure that is more suitable for human models based on their motion characteristics. In addition, this article also optimized the collision response algorithm after the system detects collisions, making the collision response process faster. Through experimental analysis, this approach has a good effect in addressing the problem of model penetration caused by continuously changing the model posture in shooting games; it also avoids the game fairness problem caused by model penetration and improves the realism of the game's virtual environment.
{"title":"An Improved Fast Collision Detection Algorithm for Human Models Based on Hybrid Bounding Boxes","authors":"Xuezhi Yue, Ye Fan, Yuan Zeng, Weitao Fan, Luhui Zhou","doi":"10.4018/ijcini.345655","DOIUrl":"https://doi.org/10.4018/ijcini.345655","url":null,"abstract":"This article analyzes the commonly used collision detection methods in game engines and designs a hybrid bounding box structure that is more suitable for human models based on their motion characteristics. In addition, this article also optimized the collision response algorithm after the system detects collisions, making the collision response process faster. Through experimental analysis, this approach has a good effect in addressing the problem of model penetration caused by continuously changing the model posture in shooting games; it also avoids the game fairness problem caused by model penetration and improves the realism of the game's virtual environment.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141828243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The environment of the epidemic is a life changer for everyone, and specifically for the educational system is progress. In order to overcome the challenges in the educational system, this study proposes a new intelligent approach to teaching and learning. Teaching plays a vital role in higher education and can be advanced by utilising certain tools and technologies such as internet enabled mobile applications, automated scheduling of courses, assessment etc. The Emergency Distance Learning Methodology (EDLM) is proposed to improve the teaching-learning process by implementing artificial intelligence. Comparing the proposed system with the existing qualitative response analysis methods, it is observed that the proposed system provides 98.56% accuracy over the existing models. This study aims to assess the optimisation of teaching and learning models in higher education.
{"title":"The Influence of Artificial Intelligence Technology on the Optimization of the Teaching Model of Higher Education in the Context of the Pandemic","authors":"Kaiyu Xiang, Yuanyuan Zhu","doi":"10.4018/ijcini.343519","DOIUrl":"https://doi.org/10.4018/ijcini.343519","url":null,"abstract":"The environment of the epidemic is a life changer for everyone, and specifically for the educational system is progress. In order to overcome the challenges in the educational system, this study proposes a new intelligent approach to teaching and learning. Teaching plays a vital role in higher education and can be advanced by utilising certain tools and technologies such as internet enabled mobile applications, automated scheduling of courses, assessment etc. The Emergency Distance Learning Methodology (EDLM) is proposed to improve the teaching-learning process by implementing artificial intelligence. Comparing the proposed system with the existing qualitative response analysis methods, it is observed that the proposed system provides 98.56% accuracy over the existing models. This study aims to assess the optimisation of teaching and learning models in higher education.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140972543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
At the oriented object detection in aerial remote sensing images, the perceptual field boundaries of ordinary convolutional kernels are often not parallel to the boundaries of the objects to be detected, affecting the model precision. Therefore, an object detection model (DCN-BBAV) that fuses deformable convolution networks (DCNs) and box boundary-aware vectors (BBAVs) is proposed. Firstly, a BBAV is used as the baseline, replacing the normal convolution kernels in the backbone network with deformable convolution kernels. Then, the spatial attention module (SAM) and channel attention mechanism (CAM) are used to enhance the feature extraction ability for a DCN. Finally, the dot product of the included angles of four adjacent vectors are added to the loss function of the rotation frame parameter, improving the regression precision of the boundary vector. The DCN-BBAV model demonstrates notable performance with a 77.30% mean average precision (mAP) on the DOTA dataset. Additionally, it outperforms other advanced rotating frame object detection methods, achieving impressive results of 90.52% mAP on VOC07 and 96.67% mAP on VOC12 for HRSC2016.
{"title":"Fusing DCN and BBAV for Remote Sensing Image Object Detection","authors":"Honghuan Chen, Keming Wang","doi":"10.4018/ijcini.335496","DOIUrl":"https://doi.org/10.4018/ijcini.335496","url":null,"abstract":"At the oriented object detection in aerial remote sensing images, the perceptual field boundaries of ordinary convolutional kernels are often not parallel to the boundaries of the objects to be detected, affecting the model precision. Therefore, an object detection model (DCN-BBAV) that fuses deformable convolution networks (DCNs) and box boundary-aware vectors (BBAVs) is proposed. Firstly, a BBAV is used as the baseline, replacing the normal convolution kernels in the backbone network with deformable convolution kernels. Then, the spatial attention module (SAM) and channel attention mechanism (CAM) are used to enhance the feature extraction ability for a DCN. Finally, the dot product of the included angles of four adjacent vectors are added to the loss function of the rotation frame parameter, improving the regression precision of the boundary vector. The DCN-BBAV model demonstrates notable performance with a 77.30% mean average precision (mAP) on the DOTA dataset. Additionally, it outperforms other advanced rotating frame object detection methods, achieving impressive results of 90.52% mAP on VOC07 and 96.67% mAP on VOC12 for HRSC2016.","PeriodicalId":509295,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139448612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：