Bruno Lacerda, Anna Gautier, Alex Rutherford, A. Stephens, Charlie Street, N. Hawes
In this overview paper, we present the work of the Goal-Oriented Long-Lived Systems Lab on multi-robot systems. We address multi-robot systems from a decision-making under uncertainty perspective, proposing approaches that explicitly reason about the inherent uncertainty of action execution, and how such stochasticity affects multi-robot coordination. To develop effective decision-making approaches, we take a special focus on (i) temporal uncertainty, in particular of action execution; (ii) the ability to provide rich guarantees of performance, both at a local (robot) level and at a global (team) level; and (iii) scaling up to systems with real-world impact. We summarise several pieces of work and highlight how they address the challenges above, and also hint at future research directions.
{"title":"Decision-making under uncertainty for multi-robot systems","authors":"Bruno Lacerda, Anna Gautier, Alex Rutherford, A. Stephens, Charlie Street, N. Hawes","doi":"10.3233/aic-220118","DOIUrl":"https://doi.org/10.3233/aic-220118","url":null,"abstract":"In this overview paper, we present the work of the Goal-Oriented Long-Lived Systems Lab on multi-robot systems. We address multi-robot systems from a decision-making under uncertainty perspective, proposing approaches that explicitly reason about the inherent uncertainty of action execution, and how such stochasticity affects multi-robot coordination. To develop effective decision-making approaches, we take a special focus on (i) temporal uncertainty, in particular of action execution; (ii) the ability to provide rich guarantees of performance, both at a local (robot) level and at a global (team) level; and (iii) scaling up to systems with real-world impact. We summarise several pieces of work and highlight how they address the challenges above, and also hint at future research directions.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"3 1","pages":"433-441"},"PeriodicalIF":0.8,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78898551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Qingwei Tang, Pu Yan, Jie Chen, Hui Shao, Fuyu Wang, G. Wang
Person re-identification (ReID) is a crucial task in identifying pedestrians of interest across multiple surveillance camera views. ReID methods in recent years have shown that using global features or part features of the pedestrian is extremely effective, but many models do not have further design models to make more reasonable use of global and part features. A new model is proposed to use global features more rationally and extract more fine-grained part features. Specifically, our model captures global features by using a multi-scale attention global feature extraction module, and we design a new context-based adaptive part feature extraction module to consider continuity between different body parts of pedestrians. In addition, we have added additional enhancement modules to the model to enhance its performance. Experiments show that our model achieves competitive results on the Market1501, Dukemtmc-ReID, and MSMT17 datasets. The ablation experiments demonstrate the effectiveness of each module of our model. The code of our model is available at: https://github.com/davidtqw/Person-Re-Identification.
{"title":"Person re-identification based on multi-scale global feature and weight-driven part feature","authors":"Qingwei Tang, Pu Yan, Jie Chen, Hui Shao, Fuyu Wang, G. Wang","doi":"10.3233/aic-210258","DOIUrl":"https://doi.org/10.3233/aic-210258","url":null,"abstract":"Person re-identification (ReID) is a crucial task in identifying pedestrians of interest across multiple surveillance camera views. ReID methods in recent years have shown that using global features or part features of the pedestrian is extremely effective, but many models do not have further design models to make more reasonable use of global and part features. A new model is proposed to use global features more rationally and extract more fine-grained part features. Specifically, our model captures global features by using a multi-scale attention global feature extraction module, and we design a new context-based adaptive part feature extraction module to consider continuity between different body parts of pedestrians. In addition, we have added additional enhancement modules to the model to enhance its performance. Experiments show that our model achieves competitive results on the Market1501, Dukemtmc-ReID, and MSMT17 datasets. The ablation experiments demonstrate the effectiveness of each module of our model. The code of our model is available at: https://github.com/davidtqw/Person-Re-Identification.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"27 1","pages":"207-223"},"PeriodicalIF":0.8,"publicationDate":"2022-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80820904","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Although convolutional neural networks (CNNs) are leading the way in semantic segmentation, standard methods still have some flaws. First, there is feature redundancy and less discriminating feature representations. Second, the number of effective multi-scale features is limited. In this paper, we aim to solve these constraints with the proposed network that utilizes two effective pre-trained models as an encoder. We develop a cross-form attention pyramid that acquires semantically rich multi-scale information from local and global priors. A spatial-wise attention module is introduced to further enhance the segmentation findings. It highlights more discriminating regions of low-level features to focus on significant location information. We demonstrate the efficacy of the proposed network on three datasets, including IDD Lite, PASCAL VOC 2012, and CamVid. Our model achieves a mIoU score of 70.7% on the IDD Lite, 83.98% on the PASCAL VOC 2012, and 73.8% on the CamVid dataset.
{"title":"Cross-form efficient attention pyramidal network for semantic image segmentation","authors":"Anamika Maurya, S. Chand","doi":"10.3233/aic-210266","DOIUrl":"https://doi.org/10.3233/aic-210266","url":null,"abstract":"Although convolutional neural networks (CNNs) are leading the way in semantic segmentation, standard methods still have some flaws. First, there is feature redundancy and less discriminating feature representations. Second, the number of effective multi-scale features is limited. In this paper, we aim to solve these constraints with the proposed network that utilizes two effective pre-trained models as an encoder. We develop a cross-form attention pyramid that acquires semantically rich multi-scale information from local and global priors. A spatial-wise attention module is introduced to further enhance the segmentation findings. It highlights more discriminating regions of low-level features to focus on significant location information. We demonstrate the efficacy of the proposed network on three datasets, including IDD Lite, PASCAL VOC 2012, and CamVid. Our model achieves a mIoU score of 70.7% on the IDD Lite, 83.98% on the PASCAL VOC 2012, and 73.8% on the CamVid dataset.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"50 1","pages":"225-242"},"PeriodicalIF":0.8,"publicationDate":"2022-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82618966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-02DOI: 10.48550/arXiv.2208.01769
I. Ahmed, Cillian Brewitt, Ignacio Carlucho, Filippos Christianos, Mhairi Dunion, Elliot Fosong, Samuel Garcin, Shangmin Guo, Balint Gyevnar, Trevor A. McInroe, Georgios Papoudakis, Arrasy Rahman, Lukas Schafer, Massimiliano Tamborski, Giuseppe Vecchio, Cheng Wang, Stefano V. Albrecht
The development of autonomous agents which can interact with other agents to accomplish a given task is a core area of research in artificial intelligence and machine learning. Towards this goal, the Autonomous Agents Research Group develops novel machine learning algorithms for autonomous systems control, with a specific focus on deep reinforcement learning and multi-agent reinforcement learning. Research problems include scalable learning of coordinated agent policies and inter-agent communication; reasoning about the behaviours, goals, and composition of other agents from limited observations; and sample-efficient learning based on intrinsic motivation, curriculum learning, causal inference, and representation learning. This article provides a broad overview of the ongoing research portfolio of the group and discusses open problems for future directions.
{"title":"Deep Reinforcement Learning for Multi-Agent Interaction","authors":"I. Ahmed, Cillian Brewitt, Ignacio Carlucho, Filippos Christianos, Mhairi Dunion, Elliot Fosong, Samuel Garcin, Shangmin Guo, Balint Gyevnar, Trevor A. McInroe, Georgios Papoudakis, Arrasy Rahman, Lukas Schafer, Massimiliano Tamborski, Giuseppe Vecchio, Cheng Wang, Stefano V. Albrecht","doi":"10.48550/arXiv.2208.01769","DOIUrl":"https://doi.org/10.48550/arXiv.2208.01769","url":null,"abstract":"The development of autonomous agents which can interact with other agents to accomplish a given task is a core area of research in artificial intelligence and machine learning. Towards this goal, the Autonomous Agents Research Group develops novel machine learning algorithms for autonomous systems control, with a specific focus on deep reinforcement learning and multi-agent reinforcement learning. Research problems include scalable learning of coordinated agent policies and inter-agent communication; reasoning about the behaviours, goals, and composition of other agents from limited observations; and sample-efficient learning based on intrinsic motivation, curriculum learning, causal inference, and representation learning. This article provides a broad overview of the ongoing research portfolio of the group and discusses open problems for future directions.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"28 1","pages":"357-368"},"PeriodicalIF":0.8,"publicationDate":"2022-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90892681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.48550/arXiv.2208.00096
Majd Hawasly, Jonathan Sadeghi, Morris Antonello, Stefano V. Albrecht, John Redford, S. Ramamoorthy
Achieving safe and robust autonomy is the key bottleneck on the path towards broader adoption of autonomous vehicles technology. This motivates going beyond extrinsic metrics such as miles between disengagement, and calls for approaches that embody safety by design. In this paper, we address some aspects of this challenge, with emphasis on issues of motion planning and prediction. We do this through description of novel approaches taken to solving selected sub-problems within an autonomous driving stack, in the process introducing the design philosophy being adopted within Five. This includes safe-by-design planning, interpretable as well as verifiable prediction, and modelling of perception errors to enable effective sim-to-real and real-to-sim transfer within the testing pipeline of a realistic autonomous system.
{"title":"Perspectives on the System-level Design of a Safe Autonomous Driving Stack","authors":"Majd Hawasly, Jonathan Sadeghi, Morris Antonello, Stefano V. Albrecht, John Redford, S. Ramamoorthy","doi":"10.48550/arXiv.2208.00096","DOIUrl":"https://doi.org/10.48550/arXiv.2208.00096","url":null,"abstract":"Achieving safe and robust autonomy is the key bottleneck on the path towards broader adoption of autonomous vehicles technology. This motivates going beyond extrinsic metrics such as miles between disengagement, and calls for approaches that embody safety by design. In this paper, we address some aspects of this challenge, with emphasis on issues of motion planning and prediction. We do this through description of novel approaches taken to solving selected sub-problems within an autonomous driving stack, in the process introducing the design philosophy being adopted within Five. This includes safe-by-design planning, interpretable as well as verifiable prediction, and modelling of perception errors to enable effective sim-to-real and real-to-sim transfer within the testing pipeline of a realistic autonomous system.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"447 1","pages":"285-294"},"PeriodicalIF":0.8,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77847852","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Loan Ho, S. Arch-int, Erman Acar, S. Schlobach, N. Arch-int
Prioritized Datalog ± is a well-studied formalism for modelling ontological knowledge and data, and has a success story in many applications in the (Semantic) Web and in other domains. Since the information content on the Web is both inherently context-dependent and frequently updated, the occurrence of a logical inconsistency is often inevitable. This phenomenon has led the research community to develop various types of inconsistency-tolerant semantics over the last few decades. Although the study of query answering under inconsistency-tolerant semantics is well-understood, the problem of explaining query answering under such semantics took considerably less attention, especially in the scenario where the facts are prioritized. In this paper, we aim to fill this gap. More specifically, we use Dung’s abstract argumentation framework to address the problem of explaining inconsistency-tolerant query answering in Datalog ± KB where facts are prioritized, or preordered. We clarify the relationship between preferred repair semantics and various notions of extensions for argumentation frameworks. The strength of such argumentation-based approach is the explainability; users can more easily understand why different points of views are conflicting and why the query answer is entailed (or not) under different semantics. To this end we introduce the formal notion of a dialogical explanation, and show how it can be used to both explain showing why query results hold and not hold according to the known semantics in inconsistent Datalog ± knowledge bases.
{"title":"An argumentative approach for handling inconsistency in prioritized Datalog± ontologies","authors":"Loan Ho, S. Arch-int, Erman Acar, S. Schlobach, N. Arch-int","doi":"10.3233/aic-220087","DOIUrl":"https://doi.org/10.3233/aic-220087","url":null,"abstract":"Prioritized Datalog ± is a well-studied formalism for modelling ontological knowledge and data, and has a success story in many applications in the (Semantic) Web and in other domains. Since the information content on the Web is both inherently context-dependent and frequently updated, the occurrence of a logical inconsistency is often inevitable. This phenomenon has led the research community to develop various types of inconsistency-tolerant semantics over the last few decades. Although the study of query answering under inconsistency-tolerant semantics is well-understood, the problem of explaining query answering under such semantics took considerably less attention, especially in the scenario where the facts are prioritized. In this paper, we aim to fill this gap. More specifically, we use Dung’s abstract argumentation framework to address the problem of explaining inconsistency-tolerant query answering in Datalog ± KB where facts are prioritized, or preordered. We clarify the relationship between preferred repair semantics and various notions of extensions for argumentation frameworks. The strength of such argumentation-based approach is the explainability; users can more easily understand why different points of views are conflicting and why the query answer is entailed (or not) under different semantics. To this end we introduce the formal notion of a dialogical explanation, and show how it can be used to both explain showing why query results hold and not hold according to the known semantics in inconsistent Datalog ± knowledge bases.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"6 1","pages":"243-267"},"PeriodicalIF":0.8,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81898610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Printed Circuit Board (PCB) is the heart component of electronic products, and its defect detection is the basic requirement of PCB quality control in the production process. Traditional visual detection methods need artificial design features, so their detection accuracy is poor, and the rate of missed and false detection is high. To solve the above problems, this paper proposes an improved YOLOv3 (You Only Look Once) detection method for PCB plug-in solder spot defects based on the combination of the ordered probability density weighting and the attention mechanism. First, in order to obtain a higher priority priori box, the ordered probability density weighting (OWA) method is used to optimize the multiple sets of a priori boxes generated by K-means. Then, to get more effective feature information, the Squeeze-and-Excitation mechanism (SE) is added to the backbone network. In the feature detection network, the Convolutional Block Attention Module (CBAM) attention mechanism is joined, at the same time in the inspection network output layer three layer feature are fusions. Finally, in order to accelerate the convergence speed of model and improve the accuracy of the model, the network loss function was improved by using the generalized joint generalized intersection over union (GIoU), and the COCO data model was applied to PCB solder spot defect training by transfer learning method. After testing, the average detection accuracy of improved network is improved from 84.35% to 96.69%, and the improved network has better convergence than the original network. The study shows that the improved method based on YOLOv3 is more suitable for industrial application of PCB plug-in solder spot defect detection.
印刷电路板(PCB)是电子产品的心脏部件，其缺陷检测是生产过程中PCB质量控制的基本要求。传统的视觉检测方法需要人工设计特征，检测精度较差，漏检率和误检率较高。针对上述问题，本文提出了一种基于有序概率密度加权和注意机制相结合的PCB插件焊点缺陷改进YOLOv3 (You Only Look Once)检测方法。首先，为了获得优先级更高的先验盒，采用有序概率密度加权(OWA)方法对K-means生成的多组先验盒进行优化。然后，为了获得更有效的特征信息，在骨干网中加入了挤压激励机制(SE)。在特征检测网络中加入了卷积块注意模块(CBAM)的注意机制，同时在检测网络输出层中对三层特征进行融合。最后，为了加快模型的收敛速度，提高模型的精度，采用广义联合广义交联(GIoU)对网络损失函数进行改进，并将COCO数据模型应用于PCB焊点缺陷的迁移学习训练。经过测试，改进后的网络平均检测准确率由84.35%提高到96.69%，收敛性优于原网络。研究表明，基于YOLOv3的改进方法更适合PCB插件焊点缺陷检测的工业应用。
{"title":"Improved YOLOv3 detection method for PCB plug-in solder joint defects based on ordered probability density weighting and attention mechanism","authors":"Zheng Wang, Wenbin Chen, Taifu Li, Shaolin Zhang, Rui Xiong","doi":"10.3233/aic-210245","DOIUrl":"https://doi.org/10.3233/aic-210245","url":null,"abstract":"Printed Circuit Board (PCB) is the heart component of electronic products, and its defect detection is the basic requirement of PCB quality control in the production process. Traditional visual detection methods need artificial design features, so their detection accuracy is poor, and the rate of missed and false detection is high. To solve the above problems, this paper proposes an improved YOLOv3 (You Only Look Once) detection method for PCB plug-in solder spot defects based on the combination of the ordered probability density weighting and the attention mechanism. First, in order to obtain a higher priority priori box, the ordered probability density weighting (OWA) method is used to optimize the multiple sets of a priori boxes generated by K-means. Then, to get more effective feature information, the Squeeze-and-Excitation mechanism (SE) is added to the backbone network. In the feature detection network, the Convolutional Block Attention Module (CBAM) attention mechanism is joined, at the same time in the inspection network output layer three layer feature are fusions. Finally, in order to accelerate the convergence speed of model and improve the accuracy of the model, the network loss function was improved by using the generalized joint generalized intersection over union (GIoU), and the COCO data model was applied to PCB solder spot defect training by transfer learning method. After testing, the average detection accuracy of improved network is improved from 84.35% to 96.69%, and the improved network has better convergence than the original network. The study shows that the improved method based on YOLOv3 is more suitable for industrial application of PCB plug-in solder spot defect detection.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"58 1","pages":"171-186"},"PeriodicalIF":0.8,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90254002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
AI Communications is the official partner journal of the European Association for Artificial Intelligence (EurAI), as reflected among others in its subtitle: the European Journal on Artificial Intelligence. EurAI is a society of societies; that is, the members of EurAI are national AI societies from all across Europe. To strengthen the connection between AI Communications and EurAI, in July 2021 we have invited each of these EurAI member societies to nominate one paper, reflecting the best research from within their society during the preceding year. Each society was free to select the criteria for their nominations. Overall, the call led to expressions of interest from 10 societies, 7 of which resulted in a submission for this special issue. All the submissions went through a fast-tracked version of the normal peer review process, which led to the acceptance of the 5 papers in this special issue. These papers, and their nominating societies, are: these works of research in
{"title":"Highlights of AI Research in Europe","authors":"S. Schockaert, R. Peñaloza","doi":"10.3233/aic-229002","DOIUrl":"https://doi.org/10.3233/aic-229002","url":null,"abstract":"AI Communications is the official partner journal of the European Association for Artificial Intelligence (EurAI), as reflected among others in its subtitle: the European Journal on Artificial Intelligence. EurAI is a society of societies; that is, the members of EurAI are national AI societies from all across Europe. To strengthen the connection between AI Communications and EurAI, in July 2021 we have invited each of these EurAI member societies to nominate one paper, reflecting the best research from within their society during the preceding year. Each society was free to select the criteria for their nominations. Overall, the call led to expressions of interest from 10 societies, 7 of which resulted in a submission for this special issue. All the submissions went through a fast-tracked version of the normal peer review process, which led to the acceptance of the 5 papers in this special issue. These papers, and their nominating societies, are: these works of research in","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"32 1","pages":"45"},"PeriodicalIF":0.8,"publicationDate":"2022-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85035894","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
3D skeleton data has been widely used in action recognition as the skeleton-based method has achieved good performance in complex dynamic environments. The rise of spatio-temporal graph convolutions has attracted much attention to use graph convolution to extract spatial and temporal features together in the field of skeleton-based action recognition. However, due to the huge difference in the focus of spatial and temporal features, it is difficult to improve the efficiency of extracting the spatiotemporal features. In this paper, we propose a channel attention and multi-scale neural network (CA-MSN) for skeleton-based action recognition with a series of spatio-temporal extraction modules. We exploit the relationship of body joints hierarchically through two modules, i.e., a spatial module which uses the residual GCN network with the channel attention block to extract the high-level spatial features, and a temporal module which uses the multi-scale TCN network to extract the temporal features at different scales. We perform extensive experiments on both the NTU-RGBD60 and NTU-RGBD120 datasets to verify the effectiveness of our network. The comparison results show that our method achieves the state-of-the-art performance with the competitive computing speed. In order to test the application effect of our CA-MSN model, we design a multi-task tandem network consisting of 2D pose estimation, 2D to 3D pose regression and skeleton action recognition model. The end-to-end (RGB video-to-action type) recognition effect is demonstrated. The code is available at https://github.com/Rh-Dang/CA-MSN-action-recognition.git.
{"title":"Channel attention and multi-scale graph neural networks for skeleton-based action recognition","authors":"Ronghao Dang, Chengju Liu, Meilin Liu, Qi Chen","doi":"10.3233/aic-210250","DOIUrl":"https://doi.org/10.3233/aic-210250","url":null,"abstract":"3D skeleton data has been widely used in action recognition as the skeleton-based method has achieved good performance in complex dynamic environments. The rise of spatio-temporal graph convolutions has attracted much attention to use graph convolution to extract spatial and temporal features together in the field of skeleton-based action recognition. However, due to the huge difference in the focus of spatial and temporal features, it is difficult to improve the efficiency of extracting the spatiotemporal features. In this paper, we propose a channel attention and multi-scale neural network (CA-MSN) for skeleton-based action recognition with a series of spatio-temporal extraction modules. We exploit the relationship of body joints hierarchically through two modules, i.e., a spatial module which uses the residual GCN network with the channel attention block to extract the high-level spatial features, and a temporal module which uses the multi-scale TCN network to extract the temporal features at different scales. We perform extensive experiments on both the NTU-RGBD60 and NTU-RGBD120 datasets to verify the effectiveness of our network. The comparison results show that our method achieves the state-of-the-art performance with the competitive computing speed. In order to test the application effect of our CA-MSN model, we design a multi-task tandem network consisting of 2D pose estimation, 2D to 3D pose regression and skeleton action recognition model. The end-to-end (RGB video-to-action type) recognition effect is demonstrated. The code is available at https://github.com/Rh-Dang/CA-MSN-action-recognition.git.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"106 1","pages":"187-205"},"PeriodicalIF":0.8,"publicationDate":"2022-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82457126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The purpose of the salient object detection (SOD) task is to suppress the background noise and segment the salient foreground regions. Some previous methods considered the strategies of background suppression and multi-level feature fusion. Other methods encountered the problem that single-scale convolution features are difficult to capture the correct object size. This paper reconsiders the above problems and proposes a comprehensive solution to achieve SOD for improving the detection performance and ensuring relatively fewer parameters. First, it is difficult to achieve a better refinement effect through only one refinement operation. To this end, a multi-scale denoising module (MSDM) and multi-pooling refinement module (MPRM) are proposed to jointly complete the refinement task of multi-level features. Besides, it is difficult to fully integrate complementary features through only one feature integration operation. Mutual learning module (MLM) is proposed to preliminarily integrate multi-level features. To reduce information redundancy, multi-attention (MA) mechanism is used to assist further integration. The proposed algorithm is called multiple refinement and integration network (MRINet). Experimental results on five benchmark datasets show that MRINet outperforms state-of-the-art methods on multiple evaluation metrics. Moreover, our ResNet-based algorithm only contains 25.202 million parameters, which is less than other ResNet-based algorithms and can run at over 37 fps on a single GPU. The code will be available at https://github.com/dc3234/MRINet.
{"title":"Multiple refinement and integration network for Salient Object Detection","authors":"Chao Dai, Chen Pan, W. He, Hanqi Sun","doi":"10.3233/aic-210273","DOIUrl":"https://doi.org/10.3233/aic-210273","url":null,"abstract":"The purpose of the salient object detection (SOD) task is to suppress the background noise and segment the salient foreground regions. Some previous methods considered the strategies of background suppression and multi-level feature fusion. Other methods encountered the problem that single-scale convolution features are difficult to capture the correct object size. This paper reconsiders the above problems and proposes a comprehensive solution to achieve SOD for improving the detection performance and ensuring relatively fewer parameters. First, it is difficult to achieve a better refinement effect through only one refinement operation. To this end, a multi-scale denoising module (MSDM) and multi-pooling refinement module (MPRM) are proposed to jointly complete the refinement task of multi-level features. Besides, it is difficult to fully integrate complementary features through only one feature integration operation. Mutual learning module (MLM) is proposed to preliminarily integrate multi-level features. To reduce information redundancy, multi-attention (MA) mechanism is used to assist further integration. The proposed algorithm is called multiple refinement and integration network (MRINet). Experimental results on five benchmark datasets show that MRINet outperforms state-of-the-art methods on multiple evaluation metrics. Moreover, our ResNet-based algorithm only contains 25.202 million parameters, which is less than other ResNet-based algorithms and can run at over 37 fps on a single GPU. The code will be available at https://github.com/dc3234/MRINet.","PeriodicalId":50835,"journal":{"name":"AI Communications","volume":"247 1","pages":"31-44"},"PeriodicalIF":0.8,"publicationDate":"2022-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76756365","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","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：