检索结果-内蒙古大学图书馆

32nd ACM International Conference on Multimedia, MM 2024

作者： Chen, Jiankang Deng, Ling Gan, Zhiyong Zheng, Wei-Shi Wang, Ruixuan School of Computer Science and Engineering Sun Yat-sen University Guangdong Guangzhou China Peng Cheng Laboratory Guangdong Shenzhen China Key Laboratory of Machine Intelligence and Advanced Computing Guangdong Guangzhou China China United Network Communications Corporation Limited Guangdong Branch Guangdong Guangzhou China School of Computer Science and Engineering Key Laboratory of Machine Intelligence and Advanced Computing Sun Yat-sen University Guangdong Guangzhou China

ISBN: (纸本)9798400706868

Out-of-Distribution (OOD) detection is crucial when deploying machine learning models in open-world applications. The core challenge in OOD detection is mitigating the model's overconfidence on OOD data. While recent methods using auxiliary outlier datasets or synthesizing outlier features have shown promising OOD detection performance, they are limited due to costly data collection or simplified assumptions. In this paper, we propose a novel OOD detection framework FodFoM that innovatively combines multiple foundation models to generate two types of challenging fake outlier images for classifier training. The first type is based on BLIP-2's image captioning capability, CLIP's vision-language knowledge, and Stable Diffusion's image generation ability. Jointly utilizing these foundation models constructs fake outlier images which are semantically similar to but different from in-distribution (ID) images. For the second type, GroundingDINO's object detection ability is utilized to help construct pure background images by blurring foreground ID objects in ID images. The proposed framework can be flexibly combined with multiple existing OOD detection methods. Extensive empirical evaluations show that image classifiers with the help of constructed fake images can more accurately differentiate real OOD image from ID ones. New state-of-the-art OOD detection performance is achieved on multiple benchmarks. The code is available at https://***/Cverchen/ACMMM2024-FodFoM. © 2024 ACM.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Improving Multimodal Rumor Detection via Dynamic Graph Modeling 27th

Improving Multimodal Rumor Detection via Dynamic Graph Model...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Wu, Xinyu Hu, Xiaoxu Qin, Xugong Zhang, Peng Zeng, Gangyan Guo, Yu Zhao, Runbo Huang, Xinjian School of Cyber Science and Engineering Nanjing University of Science and Technology Nanjing China National Computer Network Response Technical Team/Coordination Center Technology Beijing China Laboratory for Advanced Computing and Intelligence Engineering Wuxi China

ISBN: (纸本)9783031784552

Due to the proliferation of rumors in social networks, automatic rumor detection has evoked increasing attention in recent years. Despite great progress achieved by exploiting multimodal features, existing works suffer from false discrimination issues due to insufficient multimodal modeling, mainly from two aspects: 1) neglect of the dynamicity of social networks. 2) misaligned multimodal features. To alleviate the issues, we propose DGM, Dynamic Graph Modeling for rumor detection. Firstly, dynamic graph attention is devised to exploit message propagation’s structural and temporal features. Secondly, we propose a modality-shared adapter to learn better multimodal representation. Thirdly, well-aligned visual-textual features are introduced to achieve better multimodality alignment and fusion, together with cross-modal attention and alignment supervision. We conduct extensive experiments on two public datasets, demonstrating the effectiveness and superiority of DGM. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Loc4Plan: Locating Before Planning for Outdoor Vision and Language Navigation 24

Loc4Plan: Locating Before Planning for Outdoor Vision and La...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Tian, Huilin Meng, Jingke Zheng, Wei-Shi Li, Yuan-Ming Yan, Junkai Zhang, Yunong School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangdong Provincial Key Laboratory of Information Security Technology Guangzhou China

ISBN: (纸本)9798400706868

Vision and Language Navigation (VLN) is a challenging task that requires agents to understand instructions and navigate to the destination in a visual environment. One of the key challenges in outdoor VLN is keeping track of which part of the instruction was completed. To alleviate this problem, previous works mainly focus on grounding the natural language to the visual input, but neglecting the crucial role of the agent's spatial position information in the grounding process. In this work, we first explore the substantial effect of spatial position locating on the grounding of outdoor VLN, drawing inspiration from human navigation. In real-world navigation scenarios, before planning a path to the destination, humans typically need to figure out their current location. This observation underscores the pivotal role of spatial localization in the navigation process. In this work, we introduce a novel framework, Locating before Planning (Loc4Plan), designed to incorporate spatial perception for action planning in outdoor VLN tasks. The main idea behind Loc4Plan is to perform the spatial localization before planning a decision action based on corresponding guidance, which comprises a block-aware spatial locating (BAL) module and a spatial-aware action planning (SAP) module. Specifically, to help the agent perceive its spatial location in the environment, we propose to learn a position predictor that measures how far the agent is from the next intersection for reflecting its position, which is achieved by the BAL module. After this locating process, we propose the PSA module to associate visual observations After the locating process, we propose the SAP module to incorporate spatial information to ground the corresponding guidance and enhance the precision of action planning. Extensive experiments on the Touchdown and map2seq datasets show that the proposed Loc4Plan outperforms the SOTA methods. © 2024 ACM.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

PixelFade: Privacy-preserving Person Re-identification with Noise-guided Progressive Replacement 24

PixelFade: Privacy-preserving Person Re-identification with ...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Zhang, Delong Peng, Yi-Xing Wu, Xiao-Ming Wu, Ancong Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangdong Province Key Laboratory of Information Security Technology Guangzhou China

ISBN: (纸本)9798400706868

Online person re-identification services face privacy breaches from potential data leakage and recovery attacks, exposing cloud-stored images to malicious attackers and triggering public concern. The privacy protection of pedestrian images is crucial. Previous privacy-preserving person re-identification methods are unable to resist recovery attacks and compromise accuracy. In this paper, we propose an iterative method (PixelFade) to optimize pedestrian images into noise-like images to resist recovery attacks. We first give an in-depth study of protected images from previous privacy methods, which reveal that the chaos of protected images can disrupt the learning of recovery models. Accordingly, Specifically, we propose Noise-guided Objective Function with the feature constraints of a specific authorization model, optimizing pedestrian images to normal-distributed noise images while preserving their original identity information as per the authorization model. To solve the above non-convex optimization problem, we propose a heuristic optimization algorithm that alternately performs the Constraint Operation and the Partial Replacement Operation. This strategy not only safeguards that original pixels are replaced with noises to protect privacy, but also guides the images towards an improved optimization direction to effectively preserve discriminative features. Extensive experiments demonstrate that our PixelFade outperforms previous methods in resisting recovery attacks and Re-ID performance. © 2024 ACM.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Confidence-Guided Feature Alignment for Cloth-Changing Person Re-identification 27th

Confidence-Guided Feature Alignment for Cloth-Changing Pers...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Huang, Sirong Zheng, Huicheng School of Computer Science and Engineering Sun Yat-Sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China Guangdong Province Key Laboratory of Information Security Technology Guangzhou China

ISBN: (纸本)9783031781094

Much progress has been made in the field of person re-identification, but changes in clothing have hindered the practical application of long-term person re-identification. Cloth-changing person re-identification (CC-ReID) aims to address this problem, with the main challenge being the extraction of discriminative features unrelated to clothing. Existing methods, which mainly focus on introducing clothing-irrelevant cues such as key points, contours, and 3D shapes, require additional modules for feature extraction, resulting in increased complexity and potential inaccuracies due to dependence on the performance of external models. Few studies directly use the original RGB images to make the model constantly focus on clothing-independent information. In this paper, we propose a Confidence-Guided Feature Alignment Network (CGFA) for CC-ReID. Specifically, we design a confidence module that automatically learns to make confidence adjustments to fine-grained information, prompting the model to mine clothing-independent discriminative features without introducing other modal cues. By transferring knowledge, we encourage the model to learn discriminative identity features that are independent of clothing bias. As a result, the confidence module can be removed during the inference phase. The proposed simple but efficient method uses only RGB modality without additional cues, and can serve as a powerful baseline for CC-ReID to drive future research. Extensive experiments on the CC-ReID datasets demonstrate the effectiveness of the proposed method, which achieves state-of-the-art performance. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Reflectance Recovery Guided Learning of Illumination-Invariant Features for Person Re-Identification 18th

Reflectance Recovery Guided Learning of Illumination-Invari...

引用

18th Chinese Conference on Biometric Recognition, CCBR 2024

作者： Chen, Xianbiao Xie, Xiaohua School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Guangdong Province Key Laboratory of Information Security Technology Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China

ISBN: (纸本)9789819610679

In real-world scenarios, the changes of illumination are noticeable and can significantly impact the performance of re-identification (ReID) algorithms. However, the existing person ReID methods predominantly concentrate on tackling challenges in scenarios with minimal illumination fluctuations. The key to tackling this issue is to extract illumination-invariant features. So we propose a joint learning framework which combines the recovery of reflectance map and extraction of illumination-invariant feature. By sharing parameters, the proposed module, which can be seamlessly detached during the inference phase, contributes to a reduction in inference computation time. We also introduce an adversarial learning mechanism, utilizing illumination category and person identity to facilitate the extraction of features that are invariant to illumination. Besides, due to the lack of person ReID dataset containing images under diverse lighting conditions, we construct a real-world dataset called ICReID in which images have drastic illumination changes. Extensive experiments demonstrate the effectiveness of the proposed method, which achieves significant performance on both synthetic and real datasets. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Research on Sleep Staging Method Based on Multi-Scale Convolution and Self-Attention Mechanism 2

Research on Sleep Staging Method Based on Multi-Scale Convol...

引用

2nd International Conference on Biomedical and Intelligent Systems, IC-BIS 2023

作者： Pan, Dongdong Li, Yingying School of Computer and Artificial Intelligence Zhengzhou University China State Key Laboratory of Mathematical Engineering and Advanced Computing PLA Strategic Support Force Information Engineering University China

ISBN: (纸本)9781510666733

Sleep is a very important physiological process for humans, and sleep staging is an important basis for evaluating sleep quality and diagnosing sleep-related diseases. Manual sleep staging is time-consuming and laborious, and the staging results are easily affected by subjective factors. Therefore, the research on automatic sleep staging models has extremely high research value and clinical application value. This paper proposes a sleep staging method based on multi-scale one-dimensional convolutional neural network and self-attention mechanism. The attention mechanism mines the channel correlation and temporal correlation of different channels respectively. This paper selects the sleep data of 153 normal subjects in the public dataset Sleep-EDF for model training and testing. The average staging accuracy rate was 84.5%, and the consistency test value was 0.781, which was better than the consistency test value of 0.75 for the inter-physician score. The experimental results show that the algorithm model in this paper has a high staging effect, can make full use of sleep data, and has important research significance for assisting clinicians in rapid sleep staging. © 2023 SPIE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Learning High-Performance Spiking Neural Networks with Multi-Compartment Spiking Neurons 12th

Learning High-Performance Spiking Neural Networks with Mult...

引用

12th International Conference on Image and Graphics, ICIG 2023

作者： Li, Xinjie Tang, Jianxiong Lai, Jianhuang School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China

ISBN: (纸本)9783031463075

In recent years, spiking neural networks (SNNs) have gained significant attention due to bio-inspired working mechanism. The VGG-like and ResNet-like architectures are widely used for SNNs modeling. However, the spiking features of such architectures are generated by the layer-wise Integrate-and-Fire (IF) dynamics. The firing of deep neurons is independent of the neurons in the shallow layers. In this paper, we propose Multi-Compartment Spiking Neural Network (MC-SNN) that integrates the MPs of the shallow neurons to fire the deep neurons. Specifically, the MC-SNN is modeled by a Multi-Compartment Leaky Integrate-and-Fire (MC-LIF) neuron and an Adaptive Gating Unit (AGU). The MC-LIF neuron models the IF dynamics using the MPs from both deep and shallow layers, and the AGU adaptively scales the MPs of the MC-LIF neuron. These increase the information interaction between spiking neurons and improve the performance of SNNs. Besides, we design the Binarized Synaptic Encoder (BSE) to reduce the computation cost for the input of SNNs. Experimental results show that the MC-SNN performs well on the neuromorphic datasets, reaching 79.52% and 81.24% on CIFAR10-DVS and N-Caltech101, respectively. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

Prompt Customization for Continual Learning

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial intelligence 2025年第5期6卷 1373-1385页

作者： Dai, Yong Hong, Xiaopeng Wang, Yabin Ma, Zhiheng Jiang, Dongmei Wang, Yaowei Shenzhen Polytechnic University School of Artificial Intelligence Shenzhen518055 China Peng Cheng Laboratory Shenzhen518055 China Harbin Institute of Technology Faculty of Computing Harbin150001 China Xi’an Jiaotong University School of Electronic and Information Engineering Xiaan710049 China Shenzhen University of Advanced Technology Shenzhen518071 China Chinese Academy of Sciences Shenzhen Institutes of Advanced Technology Shenzhen518071 China Harbin Institute of Technology School of Computing Shenzhen518055 China

Contemporary continual learning approaches typically select prompts from a pool, which function as supplementary inputs to a pretrained model. However, this strategy is hindered by the inherent noise of its selection approach when handling increasing tasks. In response to these challenges, we reformulate the prompting approach for continual learning and propose the prompt customization (PC) method. PC mainly comprises a prompt generation module (PGM) and a prompt modulation module (PMM). In contrast to conventional methods that employ hard prompt selection, PGM assigns different coefficients to prompts from a fixed-sized pool of prompts and generates tailored prompts. Moreover, PMM further modulates the prompts by adaptively assigning weights according to the correlations between input data and corresponding prompts. We evaluate our method on four benchmark datasets for three diverse settings, including the class, domain, and task-agnostic incremental learning tasks. Experimental results demonstrate consistent improvement (by up to 16.2%), yielded by the proposed method, over the state-of-the-art (SOTA) techniques. The code has been released online. © 2020 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Easy and Effective! Data Augmentation for Knowledge-Aware Dialogue Generation Via Multi-Perspective Sentences Interaction

SSRN

引用

SSRN 2024年

作者： Peng, Sisi Qu, Dan Zhang, Wenlin Zhang, Hao Li, Shunhang Xu, Minchen Information Engineering University Zhengzhou China Laboratory for Advanced Computing and Intelligence Engineering Zhengzhou China

In recent years, knowledge-based dialogue generation has become a popular research topic due to its ability to generate informative and coherent responses by incorporating external knowledge into models. However, acquiring high-quality knowledge that corresponds to the dialogue content can be difficult and requires significant time and resources. To address the issue of limited dialogue data, most research efforts focus on data augmentation to obtain more training data. Unfortunately, these approaches do not consider knowledge augmentation, resulting in limited diversity of input data and only achieving improvements on a single metric. Real-life conversations are diverse, with repetitions, reversals, and interruptions, which present a higher demand for data diversity. In this paper, we propose a simple and effective data augmentation method called Multi-perspective Sentence Interaction to enhance the connections between sentences from different perspectives. By analyzing the target responses from multiple dialogue perspectives, we gain a better understanding of the relationships between dialogue sentences, thereby achieving an expanded knowledge-based dialogue data. We conduct experiments on multiple knowledge-based dialogue datasets and different models, and the results demonstrate that our method significantly improves the quality of model generation. Specifically, we achieved a 3.5% increase in reply accuracy and a 0.1506 improvement in diversity (DIST-2). Furthermore, the accuracy of knowledge selection improved by 19.04%, and the model perplexity decreased by 31.48%. © 2024, The Authors. All rights reserved.

关键词： Knowledge based systems

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：