检索结果-内蒙古大学图书馆

International Joint Conference on Neural Networks (IJCNN)

作者： Yuhan Yao Jiawei Feng Ancong Wu Jiangqun Ni Wei-Shi Zheng School of Computer Science and Engineering Sun Yat-sen University China School of Cyber Science and Technology Sun Yat-sen University Shenzhen China Department of New Networks Peng Cheng Laboratory Shenzhen China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Guangdong Key Laboratory of Information Security Technology China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

The mission of cloth-changing person re-identification (CC-ReID) is to discover cloth-invariant and identity-related cues, while traditional person ReID methods rely on appearance features that are biased to cloth-related cues. To tackle this cloth-biased problem, many CC-ReID methods introduced auxiliary body shape information to extract cloth-invariant features, such as 2D sketch images or the 3D Skinned Multi-Person Linear (SMPL) model. However, 2D auxiliary information lacks 3D spatial features, while the 3D SMPL model encounters challenges in capturing features at a finer granularity due to manually defined parameters. To extract fine-grained 3D shape features, we estimate depth maps that contain richer shape information and propose a Fine-grained Depth feature Mining and Distillation (FDMD) framework. We introduce a depth branch and design a fine-grained local feature interaction module to mine fine-grained 3D body shape knowledge from estimated depth maps by exploring the context of semantic-aware local body-part features. To integrate cloth-invariant depth knowledge into the appearance features, the fine-grained 3D shape features are transferred to an appearance branch by feature-space-aligned distillation. Extensive experiments demonstrate that FDMD can achieve state-of-the-art performance on three widely used CC-ReID benchmarks PRCC, Celeb-reID and LaST.

关键词： Solid modeling Three-dimensional displays Shape Neural networks Benchmark testing Feature extraction Data mining

来源：评论

学校读者我要写书评

暂无评论

Label Correlation Guided Feature Selection for Multi-label Learning 9th

Label Correlation Guided Feature Selection for Multi-label ...

引用

19th International Conference on advanced Data Mining and Applications, ADMA 2023

作者： Zhang, Kai Liang, Wei Cao, Peng Yang, Jinzhu Li, Weiping Zaiane, Osmar R. Computer Science and Engineering Northeastern University Shenyang China Key Laboratory of Intelligent Computing in Medical Image of Ministry of Education Northeastern University Shenyang China National Frontiers Science Center for Industrial Intelligence and Systems Optimization Shenyang110819 China School of Software and Microelectronics Peking University Beijing China Alberta Machine Intelligence Institute University of Alberta EdmontonAB Canada

ISBN: (纸本)9783031466731

Multi-label learning has received much attention due to its wide range of application domains. Multi-label data often has high-dimensional features, which brings more challenges to classification algorithms. Feature selection based on sparse learning is one of the most important approaches, which can select discriminative features to alleviate the curse of dimensionality. However, most of these methods do not consider feature redundancy and label correlation simultaneously. In this work, we propose a multi-label feature selection method that takes feature redundancy and label correlation into account. Specifically, the proposed method first divides features into groups according to feature correlation and then encourages competition within each group but relaxes competition between groups to eliminate redundant and irrelevant features. Moreover, we propose an approach to capture label correlation and exploit it to improve the feature selection. Finally, an iterative optimization algorithm is designed to obtain the feature weights for multi-label feature selection. Extensive experiments on various multi-label datasets demonstrate the superiority of the proposed method compared with some state-of-the-art multi-label feature selection methods. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

A Prompt Learning Framework with Large Language Model Augmentation for Few-shot Multi-label Intent Detection

A Prompt Learning Framework with Large Language Model Augmen...

引用

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

作者： Ning Zhuang Xiao Wei Junlei Li Xiaobao Wang Chenyang Wang Longbiao Wang Jianwu Dang Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) Shenzhen China Huiyan Technology (Tianjin) Co. Ltd Tianjin China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Intent detection (ID) is essential in spoken language understanding, especially in multi-label settings where intent labels are interdependent and diverse. Existing methods like SE-MLP and QA-FT struggle in few-shot settings, due to limited data availability and efficiency concerns. To address this, we introduce a Prompt Learning framework with large language Model Augmentation (PLMA) for few-shot multi-label ID. In this study, we make three contributions. First, PLMA integrates large language models (LLMs) with small language models (SLMs), using LLMs to enhance SLMs, with prompt learning as the core framework. Second, it leverages LLMs to improve the model’s understanding of both queries and labels by intent spans extraction and answer space expansion. Third, PLMA refines the QA-FT. A novel QA format template is designed to allow single-step training and inference for each input, improving efficiency. Experiments on NLU++ show that PLMA significantly outperforms other baselines in both in-domain and cross-domain settings, demonstrating its effectiveness in combining the strengths of both small and large models.

关键词： Training Adaptation models Accuracy Large language models Signal processing Natural language processing Acoustics Multilingual Speech processing

来源：评论

学校读者我要写书评

暂无评论

Progressive Human Motion Generation Based on Text and Few Motion Frames

arXiv

引用

arXiv 2025年

作者： Zeng, Ling-An Wu, Gaojie Wu, Ancong Hu, Jian-Fang Zheng, Wei-Shi School of Artificial Intelligence Sun Yat-sen University Guangdong Zhuhai519082 China School of Computer Science and Engineering Sun Yat-sen University Guangdong Guangzhou510275 China School of Computer Science and Engineering Sun Yat-sen University China The Guangdong Key Laboratory of Information Security Technology China The Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Peng Cheng Laboratory China

Although existing text-to-motion (T2M) methods can produce realistic human motion from text description, it is still difficult to align the generated motion with the desired postures since using text alone is insufficient for precisely describing diverse postures. To achieve more controllable generation, an intuitive way is to allow the user to input a few motion frames describing precise desired postures. Thus, we explore a new Text-Frame-to-Motion (TF2M) generation task that aims to generate motions from text and very few given frames. Intuitively, the closer a frame is to a given frame, the lower the uncertainty of this frame is when conditioned on this given frame. Hence, we propose a novel Progressive Motion Generation (PMG) method to progressively generate a motion from the frames with low uncertainty to those with high uncertainty in multiple stages. During each stage, new frames are generated by a Text-Frame Guided Generator conditioned on frame-aware semantics of the text, given frames, and frames generated in previous stages. Additionally, to alleviate the train-test gap caused by multi-stage accumulation of incorrectly generated frames during testing, we propose a Pseudo-frame Replacement Strategy for training. Experimental results show that our PMG outperforms existing T2M generation methods by a large margin with even one given frame, validating the effectiveness of our PMG. Code is available here. Copyright © 2025, The Authors. All rights reserved.

关键词： Human engineering

来源：评论

学校读者我要写书评

暂无评论

FLIP-80M: 80 Million Visual-Linguistic Pairs for Facial Language-Image Pre-Training 24

FLIP-80M: 80 Million Visual-Linguistic Pairs for Facial Lang...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Li, Yudong Hou, Xianxu Dezhi, Zheng Shen, Linlin Zhao, Zhe School of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China School of AI and Advanced Computing Xi'an Jiaotong-Liverpool University Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Tencent AI Lab Beijing China

ISBN: (纸本)9798400706868

While significant progress has been made in multi-modal learning driven by large-scale image-text datasets, there is still a noticeable gap in the availability of such datasets within the facial domain. To facilitate and advance the field of facial representation learning, we present FLIP-80M, a large-scale visual-linguistic dataset comprising over 80 million face images paired with text descriptions. FLIP-80M is constructed by leveraging the large openly available image-text-pair dataset LAION-5B and a mixed-method approach to filter face-related pairs from both visual and linguistic perspectives. Our curation process involves face detection, face caption classification, text de-noising, and synthesis-based image augmentation. As a result, FLIP-80M stands as the largest face-text dataset to date. To evaluate the potential of our dataset, we fine-tune the CLIP model using the proposed FLIP-80M, to create FLIP (Facial Language-Image Pretraining) and assess its representation capabilities across various downstream tasks. Our experiments demonstrate that our FLIP model achieves state-of-the-art results in a range of face analysis tasks, including face parsing, face alignment, and face attribute classification. The dataset and models are available at https://***/ydli-ai/FLIP. © 2024 ACM.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Enriching Multimodal Sentiment Analysis Through Textual Emotional Descriptions of Visual-Audio Content 39

Enriching Multimodal Sentiment Analysis Through Textual Emot...

引用

39th Annual AAAI Conference on Artificial intelligence, AAAI 2025

作者： Wu, Sheng He, Dongxiao Wang, Xiaobao Wang, Longbiao Dang, Jianwu School of New Media and Communication Tianjin University Tianjin China Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) Shenzhen China Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

ISBN: (纸本)157735897X

Multimodal Sentiment Analysis (MSA) stands as a critical research frontier, seeking to comprehensively unravel human emotions by amalgamating text, audio, and visual data. Yet, discerning subtle emotional nuances within audio and video expressions poses a formidable challenge, particularly when emotional polarities across various segments appear similar. In this paper, our objective is to spotlight emotion-relevant attributes of audio and visual modalities to facilitate multimodal fusion in the context of nuanced emotional shifts in visual-audio scenarios. To this end, we introduce DEVA, a progressive fusion framework founded on textual sentiment descriptions aimed at accentuating emotional features of visual-audio content. DEVA employs an Emotional Description Generator (EDG) to transmute raw audio and visual data into textualized sentiment descriptions, thereby amplifying their emotional characteristics. These descriptions are then integrated with the source data to yield richer, enhanced features. Furthermore, DEVA incorporates the Text-guided Progressive Fusion Module (TPF), leveraging varying levels of text as a core modality guide. This module progressively fuses visual-audio minor modalities to alleviate disparities between text and visual-audio modalities. Experimental results on widely used sentiment analysis benchmark datasets, including MOSI, MOSEI, and CH-SIMS, underscore significant enhancements compared to state-of-the-art models. Moreover, fine-grained emotion experiments corroborate the robust sensitivity of DEVA to subtle emotional variations. Copyright © 2025, Association for the Advancement of Artificial intelligence (***). All rights reserved.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

CORE-TEXT: IMPROVING SCENE TEXT DETECTION WITH CONTRASTIVE RELATIONAL REASONING

CORE-TEXT: IMPROVING SCENE TEXT DETECTION WITH CONTRASTIVE R...

引用

2021 IEEE International Conference on Multimedia and Expo, ICME 2021

作者： Lin, Jingyang Pan, Yingwei Lai, Rongfeng Yang, Xuehang Chao, Hongyang Yao, Ting Sun Yat-sen University Guangzhou China The Key Laboratory of Machine Intelligence and Advanced Computing Sun Yat-sen University Ministry of Education Guangzhou China JD AI Research Beijing China

ISBN: (纸本)9781665438643

Localizing text instances in natural scenes is regarded as a fundamental challenge in computer vision. Nevertheless, owing to the extremely varied aspect ratios and scales of text instances in real scenes, most conventional text detectors suffer from the sub-text problem that only localizes the fragments of text instance (i.e., sub-texts). In this work, we quantitatively analyze the sub-text problem and present a simple yet effective design, COntrastive RElation (CORE) module, to mitigate that issue. CORE first leverages a vanilla relation block to model the relations among all text proposals (subtexts of multiple text instances) and further enhances relational reasoning via instance-level sub-text discrimination in a contrastive manner. Such way naturally learns instance-aware representations of text proposals and thus facilitates scene text detection. We integrate the CORE module into a two-stage text detector of Mask R-CNN and devise our text detector CORE-Text. Extensive experiments on four benchmarks demonstrate the superiority of CORE-Text. © 2021 IEEE

关键词： Aspect ratio

来源：评论

学校读者我要写书评

暂无评论

A Neuroinspired Contrast Mechanism Enables Few-Shot Object Detection

SSRN

引用

SSRN 2023年

作者： Yang, Lingxiao Chen, Dapeng Chen, Yifei Peng, Wei 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 Huawei Technlogies Shenzhen China

Object detectors based on deep neural networks often require a large number of annotated images for training. In the most realistic scenario, only a few training data are available;thus, Few-Shot Object Detection (FSOD) has become an attractive research topic. In this problem setting, two-stage proposal-based Faster R-CNN detectors are widely used in multiple previous studies, while its performance still lags far behind the one trained with lots of data. We identify the main bottleneck of training a detector in the few-shot scenario as the insufficient number of samples for discovering robust comparison-based relationships (e.g., proposal-level and category-level relationships). Moreover, we found that existing methods explore these relationships only during the training phase. It is interesting to investigate whether we can effectively incorporate such relationships during model training and testing to improve FSOD. To this end, we borrow the idea of "contrast" from neuroscience studies, and propose a simple yet effective architecture, named Residual Contrast Faster R-CNN (RcFRCN). Our RcFRCN extends the existing state-of-the-art FSOD detector by introducing two newly designed contrast blocks, i.e., a Residual Spatial Contrast Block and a Residual Proposal Contrast Block. The former block mainly explores cross-spatial contrast information in a feature map extracted by the backbone network, and the latter block discovers contrast cues across proposal features outputted by the RPN module. As a result, these blocks store differences and commodities in different perspectives and are used in both training and test phases, thus enhancing the ability of transfer learning. Extensive experiments demonstrate the effectiveness of our RcFRCN. Moreover, our method also helps the detector to retain more previous knowledge and obtains better performance in the detection of base classes. © 2023, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Resampling Factor Estimation via Dual-Stream Convolutional Neural Network

引用

Computers, Materials & Continua 2021年第1期66卷 647-657页

作者： Shangjun Luo Junwei Luo Wei Lu Yanmei Fang Jinhua Zeng Shaopei Shi Yue Zhang School of Data and Computer Science Guangdong Province Key Laboratory of Information Security TechnologyMinistry of Education Key Laboratory of Machine Intelligence and Advanced ComputingSun Yat-sen UniversityGuangzhou510006China Academy of Forensic Science Shanghai200063China College of Information Science and Technology Jinan UniversityGuangzhou510632China Department of Computer Science University of Massachusetts LowellLowellMA 01854USA

The estimation of image resampling factors is an important problem in image *** all the resampling factor estimation methods,spectrumbased methods are one of the most widely used methods and have attracted a lot of research ***,because of inherent ambiguity,spectrum-based methods fail to discriminate upscale and downscale operations without any prior *** general,the application of resampling leaves detectable traces in both spatial domain and frequency domain of a resampled ***,the resampling process will introduce correlations between neighboring *** this case,a set of periodic pixels that are correlated to their neighbors can be found in a resampled ***,the resampled image has distinct and strong peaks on spectrum while the spectrum of original image has no clear ***,in this paper,we propose a dual-stream convolutional neural network for image resampling factors *** of the two streams is gray stream whose purpose is to extract resampling traces features directly from the rescaled *** other is frequency stream that discovers the differences of spectrum between rescaled and original *** features from two streams are then fused to construct a feature representation including the resampling traces left in spatial and frequency domain,which is later fed into softmax layer for resampling factor *** results show that the proposed method is effective on resampling factor estimation and outperforms some CNN-based methods.

关键词： Image forensics image resampling detection parameter estimation convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Focus, Distinguish, and Prompt: Unleashing CLIP for Efficient and Flexible Scene Text Retrieval 24

Focus, Distinguish, and Prompt: Unleashing CLIP for Efficien...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Zeng, Gangyan Zhang, Yuan Wei, Jin Yang, Dongbao Zhang, Peng Gao, Yiwen Qin, Xugong Zhou, Yu School of Cyber Science and Engineering Nanjing University of Science and Technology Nanjing China State Key Laboratory of Media Convergence and Communication Communication University of China Beijing China Lenovo Research Beijing China Institute of Information Engineering Chinese Academy of Sciences Beijing China Laboratory for Advanced Computing and Intelligence Engineering Wuxi China TMCC College of Computer Science Nankai University Tianjin China

ISBN: (纸本)9798400706868

Scene text retrieval aims to find all images containing the query text from an image gallery. Current efforts tend to adopt an Optical Character Recognition (OCR) pipeline, which requires complicated text detection and/or recognition processes, resulting in inefficient and inflexible retrieval. Different from them, in this work we propose to explore the intrinsic potential of Contrastive Language-Image Pre-training (CLIP) for OCR-free scene text retrieval. Through empirical analysis, we observe that the main challenges of CLIP as a text retriever are: 1) limited text perceptual scale, and 2) entangled visual-semantic concepts. To this end, a novel model termed FDP (Focus, Distinguish, and Prompt) is developed. FDP first focuses on scene text via shifting the attention to the text area and probing the hidden text knowledge, and then divides the query text into content word and function word for processing, in which a semantic-aware prompting scheme and a distracted queries assistance module are utilized. Extensive experiments show that FDP significantly enhances the inference speed while achieving better or competitive retrieval accuracy compared to existing methods. Notably, on the IIIT-STR benchmark, FDP surpasses the state-of-the-art model by 4.37% with a 4 times faster speed. Furthermore, additional experiments under phrase-level and attribute-aware scene text retrieval settings validate FDP's particular advantages in handling diverse forms of query text. The source code will be available at https://***/Gyann-z/FDP. © 2024 ACM.

关键词： Semantics

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：