检索结果-内蒙古大学图书馆

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Zhangyi Hu Bin Yang Mang Ye National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan China

ISBN: (纸本)9798331314385

Visible-Infrared Person Re-identification (VI-ReID) is a challenging cross-modal retrieval task due to significant modality differences, primarily resulting from the absence of color information in the infrared modality. The development of large foundation models like Large Language Models (LLMs) and Vision Language Models (VLMs) motivates us to explore a feasible solution to empower VI-ReID with off-the-shelf large foundation models. To this end, we propose a novel Text-enhanced VI-ReID framework driven by Large Foundation Models (TVI-LFM). The core idea is to enrich the representation of the infrared modality with textual descriptions automatically generated by VLMs. Specifically, we incorporate a pre-trained VLM to extract textual features from texts generated by VLM and augmented by LLM, and incrementally fine-tune the text encoder to minimize the domain gap between generated texts and original visual modalities. Meanwhile, to enhance the infrared modality with extracted textual representations, we leverage modality alignment capabilities of VLMs and VLM-generated feature-level filters. This enables the text model to learn complementary features from the infrared modality, ensuring the semantic structural consistency between the fusion modality and the visible modality. Furthermore, we introduce modality joint learning to align features across all modalities, ensuring that textual features maintain stable semantic representation of overall pedestrian appearance during complementary information learning. Additionally, a modality ensemble retrieval strategy is proposed to leverage complementary strengths of each query modality to improve retrieval effectiveness and robustness. Extensive experiments on three expanded VI-ReID datasets demonstrate that our method significantly improves the retrieval performance, paving the way for the utilization of large foundation models in downstream multi-modal retrieval tasks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimizing Generative AI Applications: A Comparative Study of Effective Prompting Techniques

Optimizing Generative AI Applications: A Comparative Study o...

引用

Pervasive Computing and Social Networking (ICPCSN), International Conference on

作者： Shivya Jyoti Moulik Tejpal Jothi K R Department of Software Systems School of Computer Science & Engineering Vellore Institute of Technology Vellore India Department of Computational Intelligence School of Computer Science & Engineering Vellore Institute of Technology Vellore India

ISBN: (数字)9798331535193

ISBN: (纸本)9798331535209

Generative Artificial Intelligence (GenAI) has emerged as a pivotal technology across various industries, driving advancements in automation, decision-making, and content generation. This paper investigates the efficacy of the prompt engineering methods such as zero-shot, one-shot, and few-shot prompting—in optimizing GenAI systems for diverse applications. Through a comprehensive literature review and an empirical survey involving 13 use cases such as chatbots, content creation, and medical decision support, we evaluate the performance of these prompting methods. The findings reveal that few-shot prompting excels in complex tasks, while zero-shot and one-shot prompting are more effective for simpler tasks. These insights provide practical guidance for leveraging GenAI across different domains, contributing to the advancement of AI-driven solutions.

关键词： Surveys Pervasive computing Industries Decision support systems Social networking (online) Large language models Chatbots Prompt engineering Human resource management Systematic literature review

来源：评论

学校读者我要写书评

暂无评论

KAnoCLIP: Zero-Shot Anomaly Detection through Knowledge-Driven Prompt Learning and Enhanced Cross-Modal Integration

KAnoCLIP: Zero-Shot Anomaly Detection through Knowledge-Driv...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Chengyuan Li Suyang Zhou Jieping Kong Lei Qi Hui Xue College of Software Engineering Southeast University Nanjing China School of Computer Science and Engineering Southeast University Nanjing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Zero-shot anomaly detection (ZSAD) identifies anomalies without needing training samples from the target dataset, essential for scenarios with privacy concerns or limited data. Vision-language models like CLIP show potential in ZSAD but have limitations: relying on manually crafted fixed textual descriptions or anomaly prompts is time-consuming and prone to semantic ambiguity, and CLIP struggles with pixel-level anomaly segmentation, focusing more on global semantics than local details. To address these limitations, We introduce KAnoCLIP, a novel ZSAD framework that leverages vision-language models. KAnoCLIP combines general knowledge from a Large Language Model (GPT-3.5) and fine-grained, image-specific knowledge from a Visual Question Answering system (Llama3) via Knowledge-Driven Prompt Learning (KnPL). KnPL uses a knowledge-driven (KD) loss function to create learnable anomaly prompts, removing the need for fixed text prompts and enhancing generalization. KAnoCLIP includes the CLIP visual encoder with V-V attention (CLIP-VV), Bi-Directional Cross-Attention for Multi-Level Cross-Modal Interaction (Bi-CMCI), and Conv-Adapter. These components preserve local visual semantics, improve local cross-modal fusion, and align global visual features with textual information, enhancing pixel-level anomaly detection. KAnoCLIP achieves state-of-the-art performance in ZSAD across 12 industrial and medical datasets, demonstrating superior generalization compared to existing methods.

关键词： Training Visualization Image segmentation Large language models Semantics Signal processing Question answering (information retrieval) Speech processing Anomaly detection Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Large Point-to-Gaussian Model for Image-to-3D Generation 24

Large Point-to-Gaussian Model for Image-to-3D Generation

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Lu, Longfei Gao, Huachen Dai, Tao Zha, Yaohua Hou, Zhi Wu, Junta Xia, Shu-Tao Tsinghua Shenzhen International Graduate School Tsinghua University Shenzhen China Tencent Shenzhen Hunyuan China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Peng Cheng Laboratory Shenzhen China

ISBN: (纸本)9798400706868

Recently, image-to-3D approaches have significantly advanced the generation quality and speed of 3D assets based on large reconstruction models, particularly 3D Gaussian reconstruction models. Existing large 3D Gaussian models directly map 2D image to 3D Gaussian parameters, while regressing 2D image to 3D Gaussian representations is challenging without 3D priors. In this paper, we propose a large Point-to-Gaussian model, that inputs the initial point cloud produced from large 3D diffusion model conditional on 2D image to generate the Gaussian parameters, for image-to-3D generation. The point cloud provides initial 3D geometry prior for Gaussian generation, thus significantly facilitating image-to-3D Generation. Moreover, we present the Attention mechanism, Projection mechanism, and Point feature extractor, dubbed as APP block, for fusing the image features with point cloud features. The qualitative and quantitative experiments extensively demonstrate the effectiveness of the proposed approach on GSO and Objaverse datasets, and show the proposed method achieves state-of-the-art performance. © 2024 ACM.

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

Dual Branch CNN and Transformer for Cardiac Atherosclerotic Plaque Classification

Dual Branch CNN and Transformer for Cardiac Atherosclerotic ...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Lei, Haijun Zhao, Jia Tong, Guanjie Zhang, Longjiang Su, Huaqiang Lei, Baiying Shenzhen University College of Computer Science and Software Engineering Shenzhen China Shenzhen University Jinling Hospital Medical School of Nanjing University Department of Radiology Nanjing China Shenzhen University School of Biomedical Engineering Shenzhen University Medical School Shenzhen China

ISBN: (纸本)9798350337488

Accurate classification of coronary artery plaques can provide effective assistance for the diagnosis of coronary artery disease(CAD). The task of coronary artery plaque classification remains extremely challenging due to the complex anatomical structure and background of coronary arteries. 3D convolution still has limitations in feature modeling, so this study builds a dual branch bridge network based on convolution neural network (CNN) and Transformer framework, and fused the local feature extraction ability of convolution and the global modeling ability of Transformer through the bridge communication module. By using a shift attention (SA) module at the intersection of dual branch information to utilizes minimal computational complexity to fuse feature maps from both branches. The ghost plus (GP) module was aimed at balancing the enormous computational power issues of building rich semantic information and training difficulties in 3D Transformers. The proposed method have demonstrated the effectiveness through a large number of comparative and ablation experiment. © 2023 IEEE.

关键词： Coronary artery classification Feature fusion Transformer

来源：评论

学校读者我要写书评

暂无评论

EventLens: Enhancing Visual Commonsense Reasoning by Leveraging Event-Aware Pretraining and Cross-modal Linking

EventLens: Enhancing Visual Commonsense Reasoning by Leverag...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Mingjie Ma Zhihuan Yu Yichao Ma Guohui Li Zhong Yang School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China School of Software Engineering Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Visual Commonsense Reasoning (VCR) is a cognitive task, challenging models to answer visual questions, and to explain the rationale behind their answers. While Large Language Models (LLMs) offer potential for this task, VCR’s complex scenes require specialized approaches to activate their commonsense reasoning abilities, as existing Multimodal LLMs struggle with VCR’s visual events and unique reference tags. To address these challenges, we propose EventLens, which enhances VCR through Event-Aware Pretraining and Cross-modal Linking in Supervised Fine-tuning. First, we introduce a new pretraining stage that emulates human cognitive processes to improve LLM comprehension of complex scenarios. Second, during supervised fine-tuning, we leverage reference tags to explicitly bridge RoI features with text, maintaining semantic integrity across modalities. Additionally, instruct prompts and task-specific adapters help integrate LLMs’ knowledge with new commonsense reasoning. Experimental results demonstrate competitive performance with state-of-the-art methods and ablation studies verify the effectiveness of proposed EventLens.

关键词： Bridges Visualization Cognitive processes Large language models Semantics Speech enhancement Signal processing Acoustics Commonsense reasoning Video recording

来源：评论

学校读者我要写书评

暂无评论

An efficient and balanced feature pyramid network: shared-FPN

引用

Optoelectronics Letters 2025年

作者： Tian Sulong Qin Longfei Pang Wenchao Gao Ying Zhao Dexin School of Computer Science and Engineering， Tianjin University of Technology Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology， Tianjin University of Technology

In the field of object detection， it is challenging to achieve a balance between neck complexity and accuracy. To address this issue， we propose an efficient and decoupled neck module called Shared-FPN(Shared Feature Pyramid Network). Our design aims to effectively address the trade-off between these two factors. Neck modules are crucial for effectively fusing multi-scale information in neural network models. However， traditional neck designs， with their single structure， have struggled to significantly improve detection performance. In recent years， the incorporation of neural architecture search techniques in neck design has led to an increase in model complexity. This trade-off between model complexity and detection performance highlights the need for more efficient neck design approaches. Not only does Shared-FPN not increase the number of model parameters and FLOPs， but it also can be easily ported to any of the detection models. It improves on PAFPN by using transposed convolution with a large convolution kernel as the upsampling module， designing spatial pyramidal pooling-fast downsampling(SPPFD) based on shared pooling， and designing shared convolution as the right part module. To evaluate the performance of Shared-FPN in object detection tasks， we conducted experiments on object detection datasets. The results show that Shared-FPN achieved excellent performance across all sizes. In particular， on the VOC 2012 dataset， which contains mostly large objects， Shared-FPN's mAP was improved by 11.2% compared to FPN with YOLOX-s as the detector. On the COCO dataset， which contains mostly small objects， Shared-FPN's mAP was improved by 7.8% compared to FPN and 7.1% compared to PAFPN with Faster RCNN as the detector. Shared-FPN can be easily inserted into any of the detectors for better performance in various scenarios， such as small， medium or large objects.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Bilateral Virtual Companions: The Impact of Virtual Humans' Movement and Voice Realism on User Perception and Experience in Multi-user VR Cinemas

Bilateral Virtual Companions: The Impact of Virtual Humans' ...

引用

International Conference on computer Supported Cooperative Work in Design

作者： Jingfeng Hu Ding Ding Xiangyu Xu Jinghui Zhang Jiahui Jin Fang Dong College of Software Engineering Southeast University Nanjing China School of Computer Science and Engineering Southeast University Nanjing China

ISBN: (数字)9798331513054

ISBN: (纸本)9798331513061

Despite the increasing prevalence of online social interaction, challenges such as insufficient immersion and lack of interactivity still persist. To overcome these limitations, this study developed a multi-user virtual reality (VR) cinema system with motion capture (Mocap) and multi-user VR technology. The system demonstrated strengths in overcoming physical space restrictions and saving travel costs, providing a more enriched interactive experience and fulfilling social needs under special circumstances, and facilitating metaverse applications. Furthermore, to give insight into the further design of multi-user VR cinemas, this study investigated the impact of virtual humans' (VHs') characteristics on user perception and experience of bilateral virtual companions, by evaluating the impact of movement and voice realism on immersion, social presence and intimacy. The results show that while neither movement nor voice realism significantly influences immersion, voice realism rather than movement realism significantly affects social presence and intimacy.

关键词： Costs Metaverse Federated learning Virtual reality Motion pictures Motion capture Space exploration

来源：评论

学校读者我要写书评

暂无评论

Design approach centered on the activity or the learning situations: For a learner-centered learning environment (LCLE) 14

Design approach centered on the activity or the learning sit...

引用

14th IEEE Global engineering Education Conference, EDUCON 2023

作者： Wague, Aminata Bousso, Mamadou Capus, Laurence Iba der Thiam University of Thies Sustainable Development and Society Doctoral School Thies Senegal Iba der Thiam University of Thies Department of Computer Science Thies Senegal Laval University Department of Computer Science and Software Engineering QC Canada

ISBN: (纸本)9798350399431

This article deals with the design of TELE (Technology-Enhanced Learning Environment), specifically LCLE (Learner-Centered Learning Environment). It suggests adopting the activity-centered approach in favour of the non-formal learning situation to design an LCLE. After demonstrating that this approach only considers formal learning situations, we describe the different stages of its implementation process. Subsequently, we develop a comparative table of the activity paradigm's formal and non-formal learning situations, allowing us to validate the proposition in question. © 2023 IEEE.

关键词： computer aided instruction

来源：评论

学校读者我要写书评

暂无评论

Revisiting Interpolation for Noisy Label Correction 39

Revisiting Interpolation for Noisy Label Correction

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Xu, Yuanzhuo Niu, Xiaoguang Yang, Jie Su, Ruiyi Zhang, Jian Liu, Shubo Drew, Steve School of Computer Science Wuhan University China Department of Electrical and Software Engineering University of Calgary Canada

ISBN: (纸本)157735897X

Label correction methods are popular for their simple architecture in learning with noisy labels. However, they suffer severely from false label correction and achieve subpar performance compared with state-of-the-art methods. In this paper, we revisit the label correction methods through theoretical analysis of gradient scaling and demonstrate that the sample-wise dynamic and class-wise uniformity of interpolation weight prevents memorization of the mislabeled samples. We then propose DULC, a simple yet effective label correction method that uses the normalized Jensen-Shannon divergence (JSD) metric as the interpolation weight to promote sample-wise dynamic and class-wise uniformity. Additionally, we provide theoretical evidence that sharpening predictions in label correction facilitates the memorization of true class, and we achieve it by employing the augmentation strategy along with the sharpening function. Extensive experiments on CIFAR-10, CIFAR-100, TinyImageNet, WebVision and Clothing1M datasets demonstrate substantial improvements over state-of-the-art methods. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词：

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：