In global navigation satellite system denial environment,cross-view geo-localization based on image retrieval presents an exceedingly critical visual localization solution for Unmanned Aerial Vehicle(UAV)systems.The e...In global navigation satellite system denial environment,cross-view geo-localization based on image retrieval presents an exceedingly critical visual localization solution for Unmanned Aerial Vehicle(UAV)systems.The essence of cross-view geo-localization resides in matching images containing the same geographical targets from disparate platforms,such as UAV-view and satellite-view images.However,images of the same geographical targets may suffer from occlusions and geometric distortions due to variations in the capturing platform,view,and timing.The existing methods predominantly extract features by segmenting feature maps,which overlook the holistic semantic distribution and structural information of objects,resulting in loss of image information.To address these challenges,dilated neighborhood attention Transformer is employed as the feature extraction backbone,and Multi-feature representations based on Multi-scale Hierarchical Contextual Aggregation(MMHCA)is proposed.In the proposed MMHCA method,the multiscale hierarchical contextual aggregation method is utilized to extract contextual information from local to global across various granularity levels,establishing feature associations of contextual information with global and local information in the image.Subsequently,the multi-feature representations method is utilized to obtain rich discriminative feature information,bolstering the robustness of model in scenarios characterized by positional shifts,varying distances,and scale ambiguities.Comprehensive experiments conducted on the extensively utilized University-1652 and SUES-200 benchmarks indicate that the MMHCA method surpasses the existing techniques.showing outstanding results in UAV localization and navigation.展开更多
Purpose:Strong primary health care(PHC)systems require well‐established PHC education systems to enhance the skills of general practitioners(GPs).However,the literature on the experiences of international collaborati...Purpose:Strong primary health care(PHC)systems require well‐established PHC education systems to enhance the skills of general practitioners(GPs).However,the literature on the experiences of international collaboration in primary care education in low‐and middle‐income countries remains limited.The purpose of this study was to evaluate the implementation and perceived impact of the McGill‐Tongji Blended Education Program for Teacher Leaders in General Practice(referred to as the“Tongji Program”).Methods:In 2020–2021,the McGill Department of Family Medicine(Montreal,Canada)and Tongji University School of Medicine(TUSM,Shanghai,China)jointly implemented the Tongji Program in Shanghai,China to improve the teaching capacity of PHC teachers.We conducted an exploratory longitudinal case study with a mixed methods design for the evaluation.Quantitative(QUAN)data was collected through questionnaire surveys and qualitative(QUAL)data was collected through focus group discussions.Results:The evaluation showed that learners in Tongji Program were primarily female GPs(21/22,95%)with less than 4 years of experience in teaching(16/22,73%).This program was considered a successful learning experience by most participants(19/22,86%)with higher order learning tasks such as critical thinking and problem‐solving.They also agreed that this program helped them feel more prepared to teach(21/22,95%),and developed a positive attitude toward primary care(21/22,95%).The QUAL interview revealed that both the Tongji and McGill organizers noted that TUSM showed strong leadership in organization,education,and coordination.Both students and teachers agreed that by adapting training content into contextualized delivery formats and settings,the Tongji Program successfully overcame language and technology barriers.Conclusions:Committed partnerships and contextualization were key to the success of the Tongji Program.Future research should focus on how international primary care education programs affect learners'behavior in their practice settings,and explore barriers and facilitators to change.展开更多
Based on the contextual adaptation perspective of Verschueren’s Adaptation Theory,this paper explores the Chinese translation strategies of Japanese quotation sentences in the Yang translation of The Courage of One f...Based on the contextual adaptation perspective of Verschueren’s Adaptation Theory,this paper explores the Chinese translation strategies of Japanese quotation sentences in the Yang translation of The Courage of One from the perspectives of communicative context and linguistic context.The study finds that the Chinese translation of Japanese quotation sentences involves various strategies,including retaining direct quotations,converting direct quotations into statements,transforming direct quotations into attributive+noun forms,and alternating between direct and indirect quotations.This research provides a new perspective for the Chinese translation of Japanese quotation sentences and offers theoretical support for translation practices in cross-cultural communication.展开更多
Federated learning(FL)is an intricate and privacy-preserving technique that enables distributed mobile devices to collaboratively train a machine learning model.However,in real-world FL scenarios,the training performa...Federated learning(FL)is an intricate and privacy-preserving technique that enables distributed mobile devices to collaboratively train a machine learning model.However,in real-world FL scenarios,the training performance is affected by a combination of factors such as the mobility of user devices,limited communication and computational resources,thus making the user scheduling problem crucial.To tackle this problem,we jointly consider the user mobility,communication and computational capacities,and develop a stochastic optimization problem to minimize the convergence time.Specifically,we first establish a convergence bound on the training performance based on the heterogeneity of users’data,and then leverage this bound to derive the participation rate for each user.After deriving the user-specific participation rate,we aim to minimize the training latency by optimizing user scheduling under the constraints of the energy consumption and participation rate.Afterward,we transform this optimization problem to the contextual multi-armed bandit framework based on the Lyapunov method and solve it with the submodular reward enhanced linear upper confidence bound(SR-linUCB)algorithm.Experimental results demonstrate the superiority of our proposed algorithm on the training performance and time consumption compared with stateof-the-art algorithms for both independent and identically distributed(IID)and non-IID settings.展开更多
The primary challenge in weakly supervised semantic segmentation is effectively leveraging weak annotations while minimizing the performance gap compared to fully supervised methods.End-to-end model designs have gaine...The primary challenge in weakly supervised semantic segmentation is effectively leveraging weak annotations while minimizing the performance gap compared to fully supervised methods.End-to-end model designs have gained significant attention for improving training efficiency.Most current algorithms rely on Convolutional Neural Networks(CNNs)for feature extraction.Although CNNs are proficient at capturing local features,they often struggle with global context,leading to incomplete and false Class Activation Mapping(CAM).To address these limitations,this work proposes a Contextual Prototype-Based End-to-End Weakly Supervised Semantic Segmentation(CPEWS)model,which improves feature extraction by utilizing the Vision Transformer(ViT).By incorporating its intermediate feature layers to preserve semantic information,this work introduces the Intermediate Supervised Module(ISM)to supervise the final layer’s output,reducing boundary ambiguity and mitigating issues related to incomplete activation.Additionally,the Contextual Prototype Module(CPM)generates class-specific prototypes,while the proposed Prototype Discrimination Loss and Superclass Suppression Loss guide the network’s training,(LPDL)(LSSL)effectively addressing false activation without the need for extra supervision.The CPEWS model proposed in this paper achieves state-of-the-art performance in end-to-end weakly supervised semantic segmentation without additional supervision.The validation set and test set Mean Intersection over Union(MIoU)of PASCAL VOC 2012 dataset achieved 69.8%and 72.6%,respectively.Compared with ToCo(pre trained weight ImageNet-1k),MIoU on the test set is 2.1%higher.In addition,MIoU reached 41.4%on the validation set of the MS COCO 2014 dataset.展开更多
Contextual advertising is a major revenue source for today＇s companies. Keyword extraction is a key step in this kind of advertising, through which appropriate advertising keywords are extracted from Web pages so tha...Contextual advertising is a major revenue source for today＇s companies. Keyword extraction is a key step in this kind of advertising, through which appropriate advertising keywords are extracted from Web pages so that corresponding ads can be triggered. This paper describes a system that learns how to extract keywords from web pages for advertisement targeting. Firstly a text network for a single webpage is build, then PageRank is applied in the network to decide on the importance of a word, finally top-ranked words are selected as keywords of the webpage. The algorithm is tested on the corpus ofblog pages, and the experimental results prove practical and effective.展开更多
With the improvement of the national economic level,the number of vehicles is still increasing year by year.According to the statistics of National Bureau of Statics,the number is approximately up to 327 million in Ch...With the improvement of the national economic level,the number of vehicles is still increasing year by year.According to the statistics of National Bureau of Statics,the number is approximately up to 327 million in China by the end of 2018,which makes urban traffic pressure continues to rise so that the negative impact of urban traffic order is growing.Illegal parking-the common problem in the field of transportation security is urgent to be solved and traditional methods to address it are mainly based on ground loop and manual supervision,which may miss detection and cost much manpower.Due to the rapidly developing deep learning sweeping the world in recent years,object detection methods relying on background segmentation cannot meet the requirements of complex and various scenes on speed and precision.Thus,an improved Single Shot MultiBox Detector(SSD)based on deep learning is proposed in our study,we introduce attention mechanism by spatial transformer module which gives neural networks the ability to actively spatially transform feature maps and add contextual information transmission in specified layer.Finally,we found out the best connection layer in the detection model by repeated experiments especially for small objects and increased the precision by 1.5%than the baseline SSD without extra training cost.Meanwhile,we designed an illegal parking vehicle detection method by the improved SSD,reaching a high precision up to 97.3%and achieving a speed of 40FPS,superior to most of vehicle detection methods,will make contributions to relieving the negative impact of illegal parking.展开更多
A sensory stimulus can only be properly interpreted in light of the stimuli that surround it in space and time. The tilt illusion （TI） and tilt after-effect （TAE） provide good evidence that the perception of a tar...A sensory stimulus can only be properly interpreted in light of the stimuli that surround it in space and time. The tilt illusion （TI） and tilt after-effect （TAE） provide good evidence that the perception of a target depends strongly on both its spatial and temporal context. In previous studies, the TI and TAE have typically been investigated separately, so little is known about their co-effects on visual perception and information processing mechanisms. Here, we considered the influence of the spatial context and the temporal effect together and asked how center- surround context affects the TAE in foveal and para- foveal vision. Our results showed that different center-surround spatial patterns significantly affected the TAE for both foveal and para-foveal vision. In the fovea, the TAE was mainly produced by central adaptive gratings. Cross-oriented surroundings significantly inhibited the TAE, and iso-oriented surroundings slightly facilitated it; surround inhibition was much stronger than surround facilitation. In the para-fovea, the TAE was mainly decided by the surrounding patches. Likewise, a cross-oriented central patch inhibited the TAE, and an iso-oriented one facilitated it, but there was no significant difference between inhibition and facilitation. Our findings demonstrated, at the perceptual level, that our visual system adopts different mechanisms to process consistent or inconsistent central-surround orientation information and that the unequalmagnitude of surround inhibition and facilitation is vitally important for the visual system to improve the detectability or discriminability of novel or incongruent stimuli.展开更多
This study investigates how orthographic,semantic and contextual variables—including word length,concreteness,and contextual support—impact on the processing and learning of new words in a second language(L2)when fi...This study investigates how orthographic,semantic and contextual variables—including word length,concreteness,and contextual support—impact on the processing and learning of new words in a second language(L2)when first encountered during reading.Students learning English as a foreign language(EFL)were recruited to read sentences for comprehension,embedded with unfamiliar L2 words that occurred once.Immediately after this,they received a form recognition test,a meaning recall test,and a meaning recognition test.Eye-movement data showed significant effects of word length on both early and late processing of novel words,along with effects of concreteness only on late-processing eye-tracking measures.Informative contexts were read slower than neutral contexts,yet contextual support did not show any direct influence on the processing of novel words.Interestingly,initial learning of abstract words was better than concrete words in terms of form and meaning recognition.Attentional processing of novel L2 words,operationalized by total reading time,positively predicted L2 learners’recognition of new orthographic forms.Taken together,these results suggest:1)orthographic,semantic and contextual factors play distinct roles for initial processing and learning of novel words;2)online processing of novel words contributes to L2 learners’initial knowledge of unfamiliar lexical items acquired from reading.展开更多
With the popularity of 5G and the rapid development of mobile terminals,an endless stream of short video software exists.Browsing short-form mobile video in fragmented time has become the mainstream of user’s life.He...With the popularity of 5G and the rapid development of mobile terminals,an endless stream of short video software exists.Browsing short-form mobile video in fragmented time has become the mainstream of user’s life.Hence,designing an efficient short video recommendation method has become important for major network platforms to attract users and satisfy their requirements.Nevertheless,the explosive growth of data leads to the low efficiency of the algorithm,which fails to distill users’points of interest on one hand effectively.On the other hand,integrating user preferences and the content of items urgently intensify the requirements for platform recommendation.In this paper,we propose a collaborative filtering algorithm,integrating time context information and user context,which pours attention into expanding and discovering user interest.In the first place,we introduce the temporal context information into the typical collaborative filtering algorithm,and leverage the popularity penalty function to weight the similarity between recommended short videos and the historical short videos.There remains one more point.We also introduce the user situation into the traditional collaborative filtering recommendation algorithm,considering the context information of users in the generation recommendation stage,and weight the recommended short-formvideos of candidates.At last,a diverse approach is used to generate a Top-K recommendation list for users.And through a case study,we illustrate the accuracy and diversity of the proposed method.展开更多
Understanding an image goes beyond recognizing and locating the objects in it,the relationships between objects also very important in image understanding.Most previous methods have focused on recognizing local predic...Understanding an image goes beyond recognizing and locating the objects in it,the relationships between objects also very important in image understanding.Most previous methods have focused on recognizing local predictions of the relationships.But real-world image relationships often determined by the surrounding objects and other contextual information.In this work,we employ this insight to propose a novel framework to deal with the problem of visual relationship detection.The core of the framework is a relationship inference network,which is a recurrent structure designed for combining the global contextual information of the object to infer the relationship of the image.Experimental results on Stanford VRD and Visual Genome demonstrate that the proposed method achieves a good performance both in efficiency and accuracy.Finally,we demonstrate the value of visual relationship on two computer vision tasks:image retrieval and scene graph generation.展开更多
In order to improve the accuracy of biophysical parameters retrieved from remotely sensing data, a new algorithm was presented by using spatial contextual to estimate canopy variables from high-resolution remote sensi...In order to improve the accuracy of biophysical parameters retrieved from remotely sensing data, a new algorithm was presented by using spatial contextual to estimate canopy variables from high-resolution remote sensing images. The developed algorithm was used for inversion of leaf area index (LAI) from Enhanced Thematic Mapper Plus (ETM+) data by combining with optimization method to minimize cost functions. The results show that the distribution of LAI is spatially consistent with the false composition imagery from ETM+ and the accuracy of LAI is significantly improved over the results retrieved by the conventional pixelwise retrieval methods, demonstrating that this method can be reliably used to integrate spatial contextual information for inverting LAI from high-resolution remote sensing images.展开更多
Fear memory contextualization is critical for selecting adaptive behavior to survive.Contextual fear conditioning(CFC)is a classical model for elucidating related underlying neuronal circuits.The primary visual cortex...Fear memory contextualization is critical for selecting adaptive behavior to survive.Contextual fear conditioning(CFC)is a classical model for elucidating related underlying neuronal circuits.The primary visual cortex(V1)is the primary cortical region for contextual visual inputs,but its role in CFC is poorly understood.Here,our experiments demonstrated that bilateral inactivation of V1 in mice impaired CFC retrieval,and both CFC learning and extinction increased the turnover rate of axonal boutons in V1.The frequency of neuronal Ca^(2+)activity decreased after CFC learning,while CFC extinction reversed the decrease and raised it to the naïve level.Contrary to control mice,the frequency of neuronal Ca^(2+)activity increased after CFC learning in microglia-depleted mice and was maintained after CFC extinction,indicating that microglial depletion alters CFC learning and the frequency response pattern of extinction-induced Ca^(2+)activity.These findings reveal a critical role of microglia in neocortical information processing in V1,and suggest potential approaches for cellular-based manipulation of acquired fear memory.展开更多
In the linguistic field,there are disputes on the view of contextual research and the goal of universality.Some scholars believe that common features of linguistic phenomena are significant while others are in favor o...In the linguistic field,there are disputes on the view of contextual research and the goal of universality.Some scholars believe that common features of linguistic phenomena are significant while others are in favor of the perception that it is more applicable and practical to carry out contextual researches.The author tends to analyze the reality and significance of contextual researches and with the goal of universality explained,the relationship between them and further suggestion will be discussed.展开更多
基金supported by the National Natural Science Foundation of China(Nos.12072027,62103052,61603346 and 62103379)the Henan Key Laboratory of General Aviation Technology,China(No.ZHKF-230201)+3 种基金the Funding for the Open Research Project of the Rotor Aerodynamics Key Laboratory,China(No.RAL20200101)the Key Research and Development Program of Henan Province,China(Nos.241111222000 and 241111222900)the Key Science and Technology Program of Henan Province,China(No.232102220067)the Scholarship Funding from the China Scholarship Council(No.202206030079).
文摘In global navigation satellite system denial environment,cross-view geo-localization based on image retrieval presents an exceedingly critical visual localization solution for Unmanned Aerial Vehicle(UAV)systems.The essence of cross-view geo-localization resides in matching images containing the same geographical targets from disparate platforms,such as UAV-view and satellite-view images.However,images of the same geographical targets may suffer from occlusions and geometric distortions due to variations in the capturing platform,view,and timing.The existing methods predominantly extract features by segmenting feature maps,which overlook the holistic semantic distribution and structural information of objects,resulting in loss of image information.To address these challenges,dilated neighborhood attention Transformer is employed as the feature extraction backbone,and Multi-feature representations based on Multi-scale Hierarchical Contextual Aggregation(MMHCA)is proposed.In the proposed MMHCA method,the multiscale hierarchical contextual aggregation method is utilized to extract contextual information from local to global across various granularity levels,establishing feature associations of contextual information with global and local information in the image.Subsequently,the multi-feature representations method is utilized to obtain rich discriminative feature information,bolstering the robustness of model in scenarios characterized by positional shifts,varying distances,and scale ambiguities.Comprehensive experiments conducted on the extensively utilized University-1652 and SUES-200 benchmarks indicate that the MMHCA method surpasses the existing techniques.showing outstanding results in UAV localization and navigation.
基金China Scholarship Council,Grant/Award Number:202000610047McGill University+4 种基金Fonds de recherche du Québec–Santé,Grant/Award Number:315852Québec Ministry of HealthCanadian Institutes for Health Research,Strategy for Patient‐Oriented Research Mentorship ChairGlobal Health Scholars ProgramFonds de recherche du Québec‐Santé,Grant/Award Number:311200。
文摘Purpose:Strong primary health care(PHC)systems require well‐established PHC education systems to enhance the skills of general practitioners(GPs).However,the literature on the experiences of international collaboration in primary care education in low‐and middle‐income countries remains limited.The purpose of this study was to evaluate the implementation and perceived impact of the McGill‐Tongji Blended Education Program for Teacher Leaders in General Practice(referred to as the“Tongji Program”).Methods:In 2020–2021,the McGill Department of Family Medicine(Montreal,Canada)and Tongji University School of Medicine(TUSM,Shanghai,China)jointly implemented the Tongji Program in Shanghai,China to improve the teaching capacity of PHC teachers.We conducted an exploratory longitudinal case study with a mixed methods design for the evaluation.Quantitative(QUAN)data was collected through questionnaire surveys and qualitative(QUAL)data was collected through focus group discussions.Results:The evaluation showed that learners in Tongji Program were primarily female GPs(21/22,95%)with less than 4 years of experience in teaching(16/22,73%).This program was considered a successful learning experience by most participants(19/22,86%)with higher order learning tasks such as critical thinking and problem‐solving.They also agreed that this program helped them feel more prepared to teach(21/22,95%),and developed a positive attitude toward primary care(21/22,95%).The QUAL interview revealed that both the Tongji and McGill organizers noted that TUSM showed strong leadership in organization,education,and coordination.Both students and teachers agreed that by adapting training content into contextualized delivery formats and settings,the Tongji Program successfully overcame language and technology barriers.Conclusions:Committed partnerships and contextualization were key to the success of the Tongji Program.Future research should focus on how international primary care education programs affect learners'behavior in their practice settings,and explore barriers and facilitators to change.
文摘Based on the contextual adaptation perspective of Verschueren’s Adaptation Theory,this paper explores the Chinese translation strategies of Japanese quotation sentences in the Yang translation of The Courage of One from the perspectives of communicative context and linguistic context.The study finds that the Chinese translation of Japanese quotation sentences involves various strategies,including retaining direct quotations,converting direct quotations into statements,transforming direct quotations into attributive+noun forms,and alternating between direct and indirect quotations.This research provides a new perspective for the Chinese translation of Japanese quotation sentences and offers theoretical support for translation practices in cross-cultural communication.
基金supported in part by the Key Technologies R&D Program of Jiangsu under Grants BE2023022 and BE2023022-2National Natural Science Foundation of China under Grants 62471204, 62531015+2 种基金Major Natural Science Foundation of the Higher Education Institutions of Jiangsu Province under Grant 24KJA510003Shanghai Kewei 24DP1500500the Fundamental Research Funds for the Central Universities under Grant 2242025K30025
文摘Federated learning(FL)is an intricate and privacy-preserving technique that enables distributed mobile devices to collaboratively train a machine learning model.However,in real-world FL scenarios,the training performance is affected by a combination of factors such as the mobility of user devices,limited communication and computational resources,thus making the user scheduling problem crucial.To tackle this problem,we jointly consider the user mobility,communication and computational capacities,and develop a stochastic optimization problem to minimize the convergence time.Specifically,we first establish a convergence bound on the training performance based on the heterogeneity of users’data,and then leverage this bound to derive the participation rate for each user.After deriving the user-specific participation rate,we aim to minimize the training latency by optimizing user scheduling under the constraints of the energy consumption and participation rate.Afterward,we transform this optimization problem to the contextual multi-armed bandit framework based on the Lyapunov method and solve it with the submodular reward enhanced linear upper confidence bound(SR-linUCB)algorithm.Experimental results demonstrate the superiority of our proposed algorithm on the training performance and time consumption compared with stateof-the-art algorithms for both independent and identically distributed(IID)and non-IID settings.
基金funding from the following sources:National Natural Science Foundation of China(U1904119)Research Programs of Henan Science and Technology Department(232102210054)+3 种基金Chongqing Natural Science Foundation(CSTB2023NSCQ-MSX0070)Henan Province Key Research and Development Project(231111212000)Aviation Science Foundation(20230001055002)supported by Henan Center for Outstanding Overseas Scientists(GZS2022011).
文摘The primary challenge in weakly supervised semantic segmentation is effectively leveraging weak annotations while minimizing the performance gap compared to fully supervised methods.End-to-end model designs have gained significant attention for improving training efficiency.Most current algorithms rely on Convolutional Neural Networks(CNNs)for feature extraction.Although CNNs are proficient at capturing local features,they often struggle with global context,leading to incomplete and false Class Activation Mapping(CAM).To address these limitations,this work proposes a Contextual Prototype-Based End-to-End Weakly Supervised Semantic Segmentation(CPEWS)model,which improves feature extraction by utilizing the Vision Transformer(ViT).By incorporating its intermediate feature layers to preserve semantic information,this work introduces the Intermediate Supervised Module(ISM)to supervise the final layer’s output,reducing boundary ambiguity and mitigating issues related to incomplete activation.Additionally,the Contextual Prototype Module(CPM)generates class-specific prototypes,while the proposed Prototype Discrimination Loss and Superclass Suppression Loss guide the network’s training,(LPDL)(LSSL)effectively addressing false activation without the need for extra supervision.The CPEWS model proposed in this paper achieves state-of-the-art performance in end-to-end weakly supervised semantic segmentation without additional supervision.The validation set and test set Mean Intersection over Union(MIoU)of PASCAL VOC 2012 dataset achieved 69.8%and 72.6%,respectively.Compared with ToCo(pre trained weight ImageNet-1k),MIoU on the test set is 2.1%higher.In addition,MIoU reached 41.4%on the validation set of the MS COCO 2014 dataset.
基金This study is supported by Beijing Natural Science Foundation of （4092029） and the Fundamental Research Funds for the Central Universities （2009RC0217）.
文摘Contextual advertising is a major revenue source for today＇s companies. Keyword extraction is a key step in this kind of advertising, through which appropriate advertising keywords are extracted from Web pages so that corresponding ads can be triggered. This paper describes a system that learns how to extract keywords from web pages for advertisement targeting. Firstly a text network for a single webpage is build, then PageRank is applied in the network to decide on the importance of a word, finally top-ranked words are selected as keywords of the webpage. The algorithm is tested on the corpus ofblog pages, and the experimental results prove practical and effective.
基金This research has been supported by NSFC(61672495)Scientific Research Fund of Hunan Provincial Education Department(16A208)+1 种基金Project of Hunan Provincial Science and Technology Department(2017SK2405)in part by the construct program of the key discipline in Hunan Province and the CERNET Innovation Project(NGII20170715).
文摘With the improvement of the national economic level,the number of vehicles is still increasing year by year.According to the statistics of National Bureau of Statics,the number is approximately up to 327 million in China by the end of 2018,which makes urban traffic pressure continues to rise so that the negative impact of urban traffic order is growing.Illegal parking-the common problem in the field of transportation security is urgent to be solved and traditional methods to address it are mainly based on ground loop and manual supervision,which may miss detection and cost much manpower.Due to the rapidly developing deep learning sweeping the world in recent years,object detection methods relying on background segmentation cannot meet the requirements of complex and various scenes on speed and precision.Thus,an improved Single Shot MultiBox Detector(SSD)based on deep learning is proposed in our study,we introduce attention mechanism by spatial transformer module which gives neural networks the ability to actively spatially transform feature maps and add contextual information transmission in specified layer.Finally,we found out the best connection layer in the detection model by repeated experiments especially for small objects and increased the precision by 1.5%than the baseline SSD without extra training cost.Meanwhile,we designed an illegal parking vehicle detection method by the improved SSD,reaching a high precision up to 97.3%and achieving a speed of 40FPS,superior to most of vehicle detection methods,will make contributions to relieving the negative impact of illegal parking.
基金supported by the National Basic Research Development Program of China ( 2013CB329401)National High Technology Development Program (863 Program) of China (2015AA020505)+2 种基金the National Natural Science Foundation of China (91120013, 6 1375115, 3 1300912, and 31100797)the 111 Project (B12027)F undamental Research Funds for the C entral Universities of China (Z YGX2013J098)
文摘A sensory stimulus can only be properly interpreted in light of the stimuli that surround it in space and time. The tilt illusion （TI） and tilt after-effect （TAE） provide good evidence that the perception of a target depends strongly on both its spatial and temporal context. In previous studies, the TI and TAE have typically been investigated separately, so little is known about their co-effects on visual perception and information processing mechanisms. Here, we considered the influence of the spatial context and the temporal effect together and asked how center- surround context affects the TAE in foveal and para- foveal vision. Our results showed that different center-surround spatial patterns significantly affected the TAE for both foveal and para-foveal vision. In the fovea, the TAE was mainly produced by central adaptive gratings. Cross-oriented surroundings significantly inhibited the TAE, and iso-oriented surroundings slightly facilitated it; surround inhibition was much stronger than surround facilitation. In the para-fovea, the TAE was mainly decided by the surrounding patches. Likewise, a cross-oriented central patch inhibited the TAE, and an iso-oriented one facilitated it, but there was no significant difference between inhibition and facilitation. Our findings demonstrated, at the perceptual level, that our visual system adopts different mechanisms to process consistent or inconsistent central-surround orientation information and that the unequalmagnitude of surround inhibition and facilitation is vitally important for the visual system to improve the detectability or discriminability of novel or incongruent stimuli.
文摘This study investigates how orthographic,semantic and contextual variables—including word length,concreteness,and contextual support—impact on the processing and learning of new words in a second language(L2)when first encountered during reading.Students learning English as a foreign language(EFL)were recruited to read sentences for comprehension,embedded with unfamiliar L2 words that occurred once.Immediately after this,they received a form recognition test,a meaning recall test,and a meaning recognition test.Eye-movement data showed significant effects of word length on both early and late processing of novel words,along with effects of concreteness only on late-processing eye-tracking measures.Informative contexts were read slower than neutral contexts,yet contextual support did not show any direct influence on the processing of novel words.Interestingly,initial learning of abstract words was better than concrete words in terms of form and meaning recognition.Attentional processing of novel L2 words,operationalized by total reading time,positively predicted L2 learners’recognition of new orthographic forms.Taken together,these results suggest:1)orthographic,semantic and contextual factors play distinct roles for initial processing and learning of novel words;2)online processing of novel words contributes to L2 learners’initial knowledge of unfamiliar lexical items acquired from reading.
文摘With the popularity of 5G and the rapid development of mobile terminals,an endless stream of short video software exists.Browsing short-form mobile video in fragmented time has become the mainstream of user’s life.Hence,designing an efficient short video recommendation method has become important for major network platforms to attract users and satisfy their requirements.Nevertheless,the explosive growth of data leads to the low efficiency of the algorithm,which fails to distill users’points of interest on one hand effectively.On the other hand,integrating user preferences and the content of items urgently intensify the requirements for platform recommendation.In this paper,we propose a collaborative filtering algorithm,integrating time context information and user context,which pours attention into expanding and discovering user interest.In the first place,we introduce the temporal context information into the typical collaborative filtering algorithm,and leverage the popularity penalty function to weight the similarity between recommended short videos and the historical short videos.There remains one more point.We also introduce the user situation into the traditional collaborative filtering recommendation algorithm,considering the context information of users in the generation recommendation stage,and weight the recommended short-formvideos of candidates.At last,a diverse approach is used to generate a Top-K recommendation list for users.And through a case study,we illustrate the accuracy and diversity of the proposed method.
文摘Understanding an image goes beyond recognizing and locating the objects in it,the relationships between objects also very important in image understanding.Most previous methods have focused on recognizing local predictions of the relationships.But real-world image relationships often determined by the surrounding objects and other contextual information.In this work,we employ this insight to propose a novel framework to deal with the problem of visual relationship detection.The core of the framework is a relationship inference network,which is a recurrent structure designed for combining the global contextual information of the object to infer the relationship of the image.Experimental results on Stanford VRD and Visual Genome demonstrate that the proposed method achieves a good performance both in efficiency and accuracy.Finally,we demonstrate the value of visual relationship on two computer vision tasks:image retrieval and scene graph generation.
基金Project(2007CB714407) supported by the Major State Basic Research and Development Program of ChinaProject(2004DFA06300) supported by Key International Collaboration Project in Science and TechnologyProjects(40571107, 40701102) supported by the National Natural Science Foundation of China
文摘In order to improve the accuracy of biophysical parameters retrieved from remotely sensing data, a new algorithm was presented by using spatial contextual to estimate canopy variables from high-resolution remote sensing images. The developed algorithm was used for inversion of leaf area index (LAI) from Enhanced Thematic Mapper Plus (ETM+) data by combining with optimization method to minimize cost functions. The results show that the distribution of LAI is spatially consistent with the false composition imagery from ETM+ and the accuracy of LAI is significantly improved over the results retrieved by the conventional pixelwise retrieval methods, demonstrating that this method can be reliably used to integrate spatial contextual information for inverting LAI from high-resolution remote sensing images.
基金supported by the National Natural Science Foundation of China(61735016)the Natural Science Foundation of Zhejiang Province(LR20F050002)+3 种基金the Key R&D Program of Zhejiang Province(2020C03009 and 2021C03001)the Zhejiang Leading Innovation and Entrepreneurship Team(202099144)the CAMS Innovation Fund for Medical Sciences(2019-I2M-5-057)Fundamental Research Funds for the Central Universities.
文摘Fear memory contextualization is critical for selecting adaptive behavior to survive.Contextual fear conditioning(CFC)is a classical model for elucidating related underlying neuronal circuits.The primary visual cortex(V1)is the primary cortical region for contextual visual inputs,but its role in CFC is poorly understood.Here,our experiments demonstrated that bilateral inactivation of V1 in mice impaired CFC retrieval,and both CFC learning and extinction increased the turnover rate of axonal boutons in V1.The frequency of neuronal Ca^(2+)activity decreased after CFC learning,while CFC extinction reversed the decrease and raised it to the naïve level.Contrary to control mice,the frequency of neuronal Ca^(2+)activity increased after CFC learning in microglia-depleted mice and was maintained after CFC extinction,indicating that microglial depletion alters CFC learning and the frequency response pattern of extinction-induced Ca^(2+)activity.These findings reveal a critical role of microglia in neocortical information processing in V1,and suggest potential approaches for cellular-based manipulation of acquired fear memory.
文摘In the linguistic field,there are disputes on the view of contextual research and the goal of universality.Some scholars believe that common features of linguistic phenomena are significant while others are in favor of the perception that it is more applicable and practical to carry out contextual researches.The author tends to analyze the reality and significance of contextual researches and with the goal of universality explained,the relationship between them and further suggestion will be discussed.
