检索结果-内蒙古大学图书馆

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 2045-2058页

作者： Yiming Zhang Xuenan Xu Ruoyi Du Haohe Liu Yuan Dong Zheng-Hua Tan Wenwu Wang Zhanyu Ma Pattern Recognition and Intelligent System Laboratory School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing China Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China Centre for Vision Speech and Signal Processing University of Surrey Guildford U.K. Department of Electronic Systems Aalborg University Aalborg Denmark

In traditional audio captioning methods, a model is usually trained in a fully supervised manner using a human-annotated dataset containing audio-text pairs and then evaluated on the test set from the same dataset. Such methods have two limitations. First, these methods are often data-hungry and require time-consuming and expensive human annotations to obtain audio-text pairs. Second, these models often suffer from performance degradation in cross-domain scenarios, i.e., when the input audio comes from a different domain than the training set, and this issue has received little attention. To address these issues, we propose a new zero-shot method for audio captioning. Our method is built on the contrastive language-audio pre-training (CLAP) model. During training, the model reconstructs the ground-truth caption using the CLAP text encoder. In the inference stage, the model generates text descriptions from the CLAP audio embeddings of given audio inputs. To enhance the ability of the model in transitioning from text-to-text generation to audio-to-text generation, we propose to use the mixed-augmentations-based soft prompt to learn more robust latent representations, leveraging instance replacement and embedding augmentation. Additionally, we introduce the retrieval-based acoustic-aware hard prompt to improve the cross-domain performance of the model by employing the domain-agnostic label information of sound events. Extensive experiments on AudioCaps and Clotho benchmarks show the effectiveness of our proposed method, which outperforms other zero-shot audio captioning approaches for in-domain scenarios and outperforms the compared methods for cross-domain scenarios, underscoring the generalization ability of our method.

关键词： Training Decoding Semantics Data models Acoustics Electronic mail Benchmark testing Transformers Robustness Perturbation methods

来源：评论

学校读者我要写书评

暂无评论

Fire Detection Based on Flame Enhancement for Weak Fires 19th

Fire Detection Based on Flame Enhancement for Weak Fires

引用

19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024

作者： Chen, Kuan Wen, Wen Feng, Fujian Xu, Xiang Liang, Yihui School of Computer Guangdong University of Technology Guangzhou510000 China School of Computer Science Zhongshan Institute University of Electronic Science and Technology of China Zhongshan528400 China Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang550025 China

ISBN: (纸本)9789819799183

Detecting weak fire, such as overexposed and highly transparent flames, remains a significant challenge in vision-based fire detection. Convolutional Neural Network (CNN) based methods are widely used for automatic fire feature extraction, but they struggle to accurately recognize overexposed flames similar to the background and highly transparent flames that blend with the background, leading to false negative fire detection results. To address this issue, we have established a large-scale fire detection dataset with weak flames and introduce a fire detection method based on flame context enhancement. While employing a flame feature extraction module to identify and delineate the potential flame area, the proposed method primarily introduces a flame context module to capture the effective features of weak flames from this area, significantly enhancing the understanding of fires. Experimental results show that the proposed method can effectively reduce false negatives and improves the recall rate for detecting weak flames. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Premixed flames

来源：评论

学校读者我要写书评

暂无评论

SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face recognition

arXiv

引用

arXiv 2022年

作者： Zhong, Yaoyao Deng, Weihong Hu, Jiani Zhao, Dongyue Li, Xian Wen, Dongchao The Pattern Recognition and Intelligent System Laboratory School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China Co. Ltd China

Deep face recognition has achieved great success due to large-scale training databases and rapidly developing loss functions. The existing algorithms devote to realizing an ideal idea: minimizing the intra-class distance and maximizing the inter-class distance. However, they may neglect that there are also low quality training images which should not be optimized in this strict way. Considering the imperfection of training databases, we propose that intra-class and inter-class objectives can be optimized in a moderate way to mitigate overfitting problem, and further propose a novel loss function, named sigmoid-constrained hypersphere loss (SFace). Specifically, SFace imposes intra-class and inter-class constraints on a hypersphere manifold, which are controlled by two sigmoid gradient re-scale functions respectively. The sigmoid curves precisely re-scale the intra-class and inter-class gradients so that training samples can be optimized to some degree. Therefore, SFace can make a better balance between decreasing the intra-class distances for clean examples and preventing overfitting to the label noise, and contributes more robust deep face recognition models. Extensive experiments of models trained on CASIA-WebFace, VGGFace2, and MS-Celeb-1M databases, and evaluated on several face recognition benchmarks, such as LFW, MegaFace and IJB-C databases, have demonstrated the superiority of SFace. Copyright © 2022, The Authors. All rights reserved.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Adaptive Pixel Pair Generation Strategy for Image Matting Methods Based on Pixel Pair Optimization 19th

Adaptive Pixel Pair Generation Strategy for Image Matting Me...

引用

19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024

作者： Zheng, Jiamin Wen, Wen Liang, Yihui Feng, Fujian Xu, Xiang School of Computer Guangdong University of Technology Guangzhou510000 China School of Computer Science Zhongshan Institute University of Electronic Science and Technology of China Zhongshan528400 China Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang550025 China

ISBN: (纸本)9789819799183

Natural image matting plays a crucial role in numerous real-world applications. Image matting methods based on pixel pair optimization is a type of matting algorithm, which has significant advantages in parallel computation, handling images with similar foreground and background, and operating under limited computing time. However, these methods cannot provide high-quality alpha mattes for high-resolution images within limited computational resources. In this paper, we design an adaptive pixel pair generation strategy to solve the above problem. This strategy adaptively generates pixel pair for the unknown pixel according to its estimated alpha value, which promotes diversity of pixel pair and improves the efficiency of pixel pair optimization. It categorizes unknown pixels into three types and additionally incorporates nearby foreground and background pixels to generate pixel pair. Experimental results show that using our strategy can achieve high-quality alpha mattes and competitive matting performance under limited computational resources. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Oracle Character recognition using Unsupervised Discriminative Consistency Network

arXiv

引用

arXiv 2023年

作者： Wang, Mei Deng, Weihong Su, Sen The Pattern Recognition and Intelligent System Laboratory School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing100876 China

Ancient history relies on the study of ancient characters. However, real-world scanned oracle characters are difficult to collect and annotate, posing a major obstacle for oracle character recognition (OrCR). Besides, serious abrasion and inter-class similarity also make OrCR more challenging. In this paper, we propose a novel unsupervised domain adaptation method for OrCR, which enables to transfer knowledge from labeled handprinted oracle characters to unlabeled scanned data. We leverage pseudo-labeling to incorporate the semantic information into adaptation and constrain augmentation consistency to make the predictions of scanned samples consistent under different perturbations, leading to the model robustness to abrasion, stain and distortion. Simultaneously, an unsupervised transition loss is proposed to learn more discriminative features on the scanned domain by optimizing both between-class and within-class transition probability. Extensive experiments show that our approach achieves state-of-the-art result on Oracle-241 dataset and substantially outperforms the recently proposed structure-texture separation network by 15.1%. © 2023, CC BY.

关键词： Character recognition

来源：评论

学校读者我要写书评

暂无评论

Mixture Loss Function-based Classification Network for Few-shot Learning

Mixture Loss Function-based Classification Network for Few-s...

引用

Computing, Robotics and system Sciences (ICRSS), International Conference on

作者： Yansha Zhang Feng Pan Jie Wang Lin Wang College of Date Science and Information Engineering GuiZhou Minzu University Guiyang China Key Laboratory of Pattern Recognition and Intelligent System of Guizhou Province Guiyang China

Data augmentation technology can effectively solve the problem of few-shot image classification, but many approaches based on data augmentation generated sample have poor discriminability, which negatively affects the classifier and classification accuracy needs to be improved. In this regard, based on the TFFCN, a mixture loss function-based classification network for few-shot learning (MLFCN) is proposed. First of all, for the prototype tensor features with a simple mean representation, ignoring the different contributions of the sample features of support set for per class to the class prototype tensor features, so that the generated prototype tensor features can not well represent sample features of per class, affecting the discriminability of the generated samples. Introduce the improved adaptively spatial feature fusion module, it uses the convolution, considers the support set of different sample features on the prototype tensor, adaptively assign weights, then weighted and summed to obtain a more representative prototype tensor feature; secondly, aiming at the blurred classification boundary of similar categories in the generated sample features, a mixture loss function is proposed. The loss function is to add the inter-class mean squared error term to the mean square error loss, and stretches the inter-class distance of the generated tensor feature, so that the classification boundary is clearer, so as to train a better classifier. Experimental results show that the improved adaptively spatial feature fusion and the mixing loss function can augment the discriminability of the generated tensor features and further improve the classification accuracy.

关键词： Training Tensors Convolution Prototypes Mean square error methods Feature extraction Generators

来源：评论

学校读者我要写书评

暂无评论

Learning Invariant Visual Representations for Compositional Zero-Shot Learning

arXiv

引用

arXiv 2022年

作者： Zhang, Tian Liang, Kongming Du, Ruoyi Sun, Xian Ma, Zhanyu Guo, Jun Pattern Recognition and Intelligent System Laboratory School of Artificial Intelligence Beijing University of Posts and Telecommunications China Aerospace Information Research Institute Chinese Academy of Sciences China

Compositional Zero-Shot Learning (CZSL) aims to recognize novel compositions using knowledge learned from seen attribute-object compositions in the training set. Previous works mainly project an image and a composition into a common embedding space to measure their compatibility score. However, both attributes and objects share the visual representations learned above, leading the model to exploit spurious correlations and bias towards seen pairs. Instead, we reconsider CZSL as an out-of-distribution generalization problem. If an object is treated as a domain, we can learn object-invariant features to recognize the attributes attached to any object reliably. Similarly, attribute-invariant features can also be learned when recognizing the objects with attributes as domains. Specifically, we propose an invariant feature learning framework to align different domains at the representation and gradient levels to capture the intrinsic characteristics associated with the tasks. Experiments on three CZSL benchmarks demonstrate that the proposed method significantly outperforms the previous state-of-the-art. Copyright © 2022, The Authors. All rights reserved.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

MEET: A Million-Scale Dataset for Fine-Grained Geospatial Scene Classification with Zoom-Free Remote Sensing Imagery

arXiv

引用

arXiv 2025年

作者： Li, Yansheng Wu, Yuning Cheng, Gong Tao, Chao Dang, Bo Wang, Yu Zhang, Jiahao Zhang, Chuge Liu, Yiting Tang, Xu Ma, Jiayi Zhang, Yongjun Information and computing science from Shandong University Weihai China Pattern recognition and intelligent system from the Huazhong University of Science and Technology Wuhan China Computer science and technology from Wuhan University Wuhan China Xidian University Xi’an China Pattern recognition and intelligent systems from Northwestern Polytechnical University Xi’an China School of Mathematics and Statistics Huazhong University of Science and Technology Wuhan China Institute for Pattern Recognition and Artificial Intelligence Huazhong China University of Science and Technology Remote sensing science and technology from Wuhan University Wuhan China School of Remote Sensing and Information Engineering Wuhan University Wuhan China Electronic circuit and system from Xidian University Xi’an China University of Colorado at Boulder BoulderCO United States Control science and engineering from the Huazhong University of Science and Technology Wuhan China Geodesy and photography from Wuhan University Wuhan China

Accurate fine-grained geospatial scene classification using remote sensing imagery is essential for a wide range of applications. However, existing approaches often rely on manually zooming remote sensing images at different scales to create typical scene samples. This approach fails to adequately support the fixed-resolution image interpretation requirements in real-world scenarios. To address this limitation, we introduce the Million-scale finE-grained geospatial scEne classification dataseT (MEET), which contains over 1.03 million zoom-free remote sensing scene samples, manually annotated into 80 fine-grained categories. In MEET, each scene sample follows a scene-in-scene layout, where the central scene serves as the reference, and auxiliary scenes provide crucial spatial context for fine-grained classification. Moreover, to tackle the emerging challenge of scene-in-scene classification, we present the Context-Aware Transformer (CAT), a model specifically designed for this task, which adaptively fuses spatial context to accurately classify the scene samples. CAT adaptively fuses spatial context to accurately classify the scene samples by learning attentional features that capture the relationships between the center and auxiliary scenes. Based on MEET, we establish a comprehensive benchmark for fine-grained geospatial scene classification, evaluating CAT against 11 competitive baselines. The results demonstrate that CAT significantly outperforms these baselines, achieving a 1.88% higher balanced accuracy (BA) with the Swin-Large backbone, and a notable 7.87% improvement with the Swin-Huge backbone. Further experiments validate the effectiveness of each module in CAT and show the practical applicability of CAT in the urban functional zone mapping. The source code and dataset will be publicly available at https://***/project/***. Copyright © 2025, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Natural image matting based on image inpainting

Natural image matting based on image inpainting

引用

Computer Graphics, Image and Virtualization (ICCGIV), International Conference on

作者： Yuan Zhang Mian Tan Zhulian Zhou Yuan Yang Yihui Liang Fujian Feng Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang China School of Computer Science University of Electronic Science and Technology China Zhongshan China

Deep image matting is a hot problem with applications in computer vision and image processing. It has been widely used in image composition, film production and video editing etc. The current matting method based on image inpainting uses a deep neural network to inpaint the foreground target and background region to further improve the accuracy of alpha matte. However, when the trimap contains large unknown regions, the excessive inpainting of foreground and background produce a lot of redundant information, which leads to a degradation of alpha matte quality. Therefore, to address this issue, a natural image matting method based on image inpainting is designed. This method involves the refinement process for trimap, which improves the quality of the trimap, enlarges the foreground and background regions providing additional information for image matting. Extensive experimental results on the composition-1k dataset demonstrate that the presented method provide high-quality alpha mattes not only in the case that the trimap contains small unknown regions, but also in the case that the trimap contains large unknown regions.

关键词： Degradation Deep learning Computer vision Image processing Refining Neural networks Production

来源：评论

学校读者我要写书评

暂无评论

GeoPix: Multi-Modal Large Language Model for Pixel-level Image Understanding in Remote Sensing

arXiv

引用

arXiv 2025年

作者： Ou, Ruizhe Hu, Yuan Zhang, Fan Chen, Jiaxin Liu, Yu Pattern Recognition and Intelligent System Lab School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China Institute of Remote Sensing and Geographic Information Systems School of Earth and Space Sciences Peking University Beijing100871 China Peking University Ordos Research Institute of Energy Ordos017000 China

Multi-modal large language models (MLLMs) have achieved remarkable success in image- and region-level remote sensing (RS) image understanding tasks, such as image captioning, visual question answering, and visual grounding. However, existing RS MLLMs lack the pixel-level dialogue capability, which involves responding to user instructions with segmentation masks for specific instances. In this paper, we propose GeoPix, a RS MLLM that extends image understanding capabilities to the pixel level. This is achieved by equipping the MLLM with a mask predictor, which transforms visual features from the vision encoder into masks conditioned on the LLM’s segmentation token embeddings. To facilitate the segmentation of multi-scale objects in RS imagery, a class-wise learnable memory module is integrated into the mask predictor to capture and store class-wise geo-context at the instance level across the entire dataset. In addition, to address the absence of large-scale datasets for training pixel-level RS MLLMs, we construct the GeoPixInstruct dataset, comprising 65,463 images and 140,412 instances, with each instance annotated with text descriptions, bounding boxes, and masks. Furthermore, we develop a two-stage training strategy to balance the distinct requirements of text generation and masks prediction in multi-modal multi-task optimization. Extensive experiments verify the effectiveness and superiority of GeoPix in pixel-level segmentation tasks, while also maintaining competitive performance in image- and region-level benchmarks. The models, dataset, and code are publicly available at https://***/Norman-Ou/GeoPix. © 2025, CC BY.

关键词： Image segmentation

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：