检索结果-内蒙古大学图书馆

weakly supervised object detection from remote sensing images via self-attention distillation and instance-aware mining

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2023年第13期83卷 39073页

作者： Yang, Peng Zhou, Shi Wang, Linlin Yang, Guowei Nanjing Audit Univ Sch Comp Sci Nanjing 211815 Peoples R China Wuhan Univ Sch Remote Sensing & Informat Engn Wuhan 430079 Peoples R China

weakly supervised object detection (WSOD) is an effective method to train object detectors using only image-level category labels, and has been concerned in the field of remote sensing image processing due to its inexpensive annotation cost. Without object-level labels, however, WSOD methods are prone to detect discriminative object parts. Furthermore, remote sensing images often contain objects with varying scales, or dense objects positioned in extremely close proximity, which makes the detection task more challenging. To address these issues, we propose a new weakly supervised learning based detector with several elaborate designs. Our detector employs a feature pyramid network with the adaptive pooling block, which can be helpful to effectively handle objects with different scales. In the meantime, a layer-wise self-attention distillation (SAD) module is performed in our framework to improve feature representation and prevent the detector focusing on the most discriminative object parts. Moreover, an instance-aware mining (IAM) algorithm is proposed to generate more precise pseudo-labels, and thereby alleviating the problem that adjacent small targets may be detected as a single object. Since SAD utilizes the feature-level knowledge distillation, and iterative refinement with IAM is a kind of instance-level knowledge distillation, they are effectively complementary to each other. We have evaluated the proposed method on popular benchmarks, including NWPU VHR-10 and DIOR, and the experimental results show that it can locate objects with more accurate bounding boxes for remote sensing images.

关键词： Remote sensing image processing weakly supervised object detection Deep learning Knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

weakly supervised object detection With Class Prototypical Network

引用

IEEE TRANSACTIONS ON MULTIMEDIA 2023年 25卷 1868-1878页

作者： Li, Huifang Li, Yidong Cao, Yuanzhouhan Han, Yushan Jin, Yi Wei, Yunchao Beijing Jiaotong Univ Sch Comp & Informat Technol Beijing 100044 Peoples R China

In this paper, we aim to devise a new framework to compel the network to be equipped with the capability of detecting objects using image-level class labels as supervision. The challenge of such a weakly supervised setting mainly lies in how to make the network accurately understand both semantics and objectness of a given proposal without bounding box annotations. To this end, we contribute a concise framework, named Class Prototypical Network (CPNet). Concretely, our CPNet defines a set of learnable class prototypes to help classify object proposals. To endow the prototypes be not only discriminative for classes but also sensitive for proposals' objectness, we conduct both class-aware cross-attention and location-aware cross-attention between the feature embeddings of the learnable prototypes and the proposals. The learned attention scores are then used to form the proposal-level category information into the image-level one, making the entire framework be trained without any bounding box annotations. Besides, by applying these two kinds of attention mechanisms, the knowledge from both proposals' location and its class information can be successfully transferred into the corresponding prototypes. With the help of prototypes, our CPNet detects true positive object proposals. In addition, the CPNet further introduces a multi-head detection head to perform complementary training, preventing the model from falling into local discriminative parts and improving the model's performance on challenging non-rigid categories. We examine our CPNet on popular benchmarks, i.e., PASCAL VOC 2007, 2012 and MS COCO 2014. Extensive experiments show our CPNet is a simple and effective framework.

关键词： Cross-attention prototypical network weakly supervised object detection

来源：评论

学校读者我要写书评

暂无评论

weakly supervised object detection Using Proposal- and Semantic-Level Relationships

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2022年第6期44卷 3349-3363页

作者： Zhang, Dingwen Zeng, Wenyuan Yao, Jieru Han, Junwei Northwestern Polytech Univ Sch Automat Brain & Artificial Intelligence Lab Xian 710072 Peoples R China

In recent years, weakly supervised object detection has attracted great attention in the computer vision community. Although numerous deep learning-based approaches have been proposed in the past few years, such an ill-posed problem is still challenging and the learning performance is still behind the expectation. In fact, most of the existing approaches only consider the visual appearance of each proposal region but ignore to make use of the helpful context information. To this end, this paper introduces two levels of context into the weakly supervised learning framework. The first one is the proposal-level context, i.e., the relationship of the spatially adjacent proposals. The second one is the semantic-level context, i.e., the relationship of the co-occurring object categories. Therefore, the proposed weakly supervised learning framework contains not only the cognition process on the visual appearance but also the reasoning process on the proposal- and semantic-level relationships, which leads to the novel deep multiple instance reasoning framework. Specifically, built upon a conventional CNN-based network architecture, the proposed framework is equipped with two additional graph convolutional network-based reasoning models to implement object location reasoning and multi-label reasoning within an end-to-end network training procedure. Comprehensive experiments on the widely used PASCAL VOC and MS COCO benchmarks have been implemented, which demonstrate the superior capacity of the proposed approach when compared with other state-of-the-art methods and baseline models.

关键词： Cognition Proposals object detection supervised learning Semantics Task analysis Network architecture weakly supervised object detection multiple-instance learning graphical convolutional network

来源：评论

学校读者我要写书评

暂无评论

Multi-instance mining with dynamic localization for weakly supervised object detection in remote-sensing images

引用

INTERNATIONAL JOURNAL OF REMOTE SENSING 2025年第9期46卷 3487-3512页

作者： Guo, Chen Ma, Zhenyu Zhao, Yangyang Cao, Chuanshuo Jiang, Zhiguo Zhang, Haopeng Xian Modern Control Technol Res Inst Xian Peoples R China Beihang Univ Tianmushan Lab Hangzhou Peoples R China Beihang Univ Sch Astronaut Dept Aerosp Intelligent Sci & Technol 9 Nansan StShahe Univ Pk Beijing 102206 Peoples R China

weakly supervised remote-sensing image object detection is crucial in the interpretation of remote-sensing images. Current mainstream approaches often focus on selecting the highest-scoring instances for training the detection network or overlook the issue of feature inconsistency within object classes during model training, leading to missed instances. Moreover, these methods typically use the same scheme for pseudo-label selection during training, which fails to achieve dynamic object localization, thus limiting the detection performance of the model. To address these challenges, we propose a feature-consistent and spatial voting multi-instance mining framework that also implements dynamic object localization. Specifically, we design a feature consistency learning module that inputs the same image at different angles into the weakly supervised Deep detection Network (WSDDN) simultaneously, capturing objects of the same class from various angles by combining different detection results. Next, a Spatial Voting Instance Mining (SVIM) module is introduced to perform spatial voting by leveraging the scores and spatial relationships of instances. This module also applies boundary regularization to the voting results, effectively mining more reliable instances for supervision and addressing the issue of the detector focusing only on parts of the object. Finally, the object Dynamic Localization (ODL) module uses stage prediction results as pseudo-label supplements during training, which greatly enhances the localization accuracy of the detector. Extensive experiments were carried out on the challenging Northwestern Polytechnic University (NWPU) VHR-10.v2 dataset and the Optical RSI (DIOR) dataset. The experimental results demonstrated that the proposed method can significantly improve detection accuracy.

关键词： weakly supervised object detection feature consistency spatial voting dynamic localization remote-sensing images

来源：评论

学校读者我要写书评

暂无评论

Low-Shot weakly supervised object detection for Remote Sensing Images via Part Domination-Based Active Learning and Enhanced Fine-Tuning

引用

REMOTE SENSING 2025年第7期17卷 1155-1155页

作者： Liu, Peng Huang, Boxue Jin, Tingting Long, Hui Chinese Acad Sci Aerosp Informat Res Inst Beijing 100094 Peoples R China Chinese Acad Sci Key Lab Technol Geospatial Informat Proc & Applica Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Target Cognit & Applicat Technol TCAT Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Elect Elect & Commun Engn Beijing 100049 Peoples R China

In low-shot weakly supervised object detection (LS-WSOD), a small number of strong (instance-level) labels are introduced to a weakly (image-level) annotated dataset, thus balancing annotation costs and model performance. To address issues in LS-WSOD in remote sensing images (RSIs) such as part domination, context confusion, class imbalance, and noise, we propose a novel active learning strategy and an enhanced fine-tuning mechanism. Specifically, we designed a part domination-based adaptive active learning (PDAAL) strategy to discover the most informative and challenging samples for instance-level annotation. PDAAL also applies an adaptive threshold to balance sampling frequencies for long-tailed class distributions. For enhanced fine-tuning, we first developed a parameter-efficient attention for context (PAC) module that learns spatial attention relationships, mitigating context confusion and accelerating the convergence of fine-tuning. Furthermore, we present an adaptive category resampling for tuning (ACRT) mechanism for resampling strong annotation data. ACRT contributes to refining the model at different active stages, especially for under-performed classes, and to reducing the impact of noisy predictions. Experimental results on the NWPU VHR-10.v2 and DIOR datasets show that our method outperforms state-of-the-art LS-WSOD baselines by 4.5% and 3.1% in mAP, respectively, demonstrating that our framework offers an efficient solution for LS-WSOD in RSIs.

关键词： remote sensing images weakly supervised object detection active learning low-shot weakly supervised object detection

来源：评论

学校读者我要写书评

暂无评论

Towards Precise weakly supervised object detection via Interactive Contrastive Learning of Context Information

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2025年第2期9卷 1795-1804页

作者： Lai, Qi Vong, Chi-Man Shi, Sai-Qi Chen, C. L. Philip Chinese Acad Sci Shenzhen Inst Adv Technol Shenzhen 518055 Peoples R China Univ Macau Dept Comp & Informat Sci Macau 999078 Peoples R China Univ Macau Dept Elect & Comp Engn Macau 999078 Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China

weakly supervised object detection (WSOD) aims at learning precise object detectors with only image-level tags. In spite of intensive research on deep learning (DL) approaches over the past few years, there is still a significant performance gap between WSOD and fully supervised object detection. Existing WSOD methods only consider the visual appearance of each region proposal but ignore the useful context information in the image. This paper proposes an interactive end-to-end WSDO framework called JLWSOD with two innovations: i) two types of WSOD-specific context information (i.e., instance-wise correlation and semantic-wise correlation) are proposed and introduced into WSOD framework;ii) an interactive graph contrastive learning (iGCL) mechanism is designed to jointly optimize the visual appearance and context information for better WSOD performance. Specifically, the iGCL mechanism takes full advantage of the complementary interpretations of the WSOD, namely instance-wise detection and semantic-wise prediction tasks, forming a more comprehensive solution. Extensive experiments on the widely used PASCAL VOC and MS COCO benchmarks verify the superiority of JLWSOD over alternative SOTA and baseline models (improvement of 3.0%similar to 23.3% on mAP and 3.1%similar to 19.7% on CorLoc, respectively).

关键词： Context information weakly supervised object detection graph contrastive learning interactive framework

来源：评论

学校读者我要写书评

暂无评论

Pseudo-label enhancement for weakly supervised object detection using self-supervised vision transformer

引用

KNOWLEDGE-BASED SYSTEMS 2025年 311卷

作者： Yang, Kequan Wu, Yuanchen Li, Jide Yin, Chao Li, Xiaoqiang Shanghai Univ Sch Comp Engn & Sci Shanghai Peoples R China

weakly supervised object detection (WSOD) using image-level labels has gained attention in the computer vision community. Most advanced WSOD approaches generate instance-level labels based on class activation maps (CAMs), which are limited by incomplete box labels and sparse information regarding class labels. These inaccurate pseudo-labels mislead the training of the subsequent detection module, thus hampering detection performance. To address these issues, we propose a novel pseudo-label enhancement (PLE) framework fora one-stage WSOD method that consists of a CAM reassembly (CAMR) and a composite scoring module (CSM). The CAMR explores the collaboration between class priors from CAMs and the semantic-aware grouping ability of self-supervised vision transformers (ViTs). Specifically, CAMR initially propagates object localization cues from CAMs to feature maps obtained using a self-supervised ViT. Subsequently, CAMR utilizes the patch affinity of these feature maps to extract more integral object information and improve the instance localization accuracy and completeness. Subsequently, the reassembled CAMs are refined using a multi-threshold strategy to generate bounding-box pseudo-labels. The CSM replaces one-hot labels with soft-class labels that leverage localization precision and classification confidence to counter information sparseness limitations. The proposed PLE is evaluated using the PASCAL VOC and MS-COCO datasets. The experimental results demonstrate that the performance of PLE surpasses that of the advanced methods. On average, PLE improves by 7.1% mAP on the VOC 2007 test set and 6.0% mAP on the MS-COCO 2017 validation set compared with the one-stage WSOD method.

关键词： weakly supervised object detection Class activation map Self-supervised vision transformer Pseudo-labels

来源：评论

学校读者我要写书评

暂无评论

Multi-view contextual adaptation network for weakly supervised object detection in remote sensing images

引用

INTERNATIONAL JOURNAL OF REMOTE SENSING 2024年第13期45卷 4344-4366页

作者： Ye, Binfeng Zhang, Junjie Rao, Yutao Gao, Rui Zeng, Dan Shanghai Univ Shanghai Inst Adv Commun & Data Sci Key Lab Specialty Fiber Opt & Opt Access Networks Shanghai Peoples R China

weakly supervised learning plays a pivotal role in the field of object detection, i.e. weakly supervised object detection (WSOD), significantly reducing annotation costs relying on image-level labels. However, WSOD exhibits certain limitations. Typically, they tend to identify the most easily recognizable local regions within targets, posing challenges in accurately delineating the boundaries of targets. Moreover, the presence of multiple instances of the same class in adjacent locations complicates the effective distinction between multiple objects within the same category. On the other hand, the complex backgrounds and dense distribution of targets in remote sensing images (RSI) further exacerbate the difficulty of weakly supervised detection. To address the above issues, we propose a model termed the Multi-View Contextual Adaptation Network (VCANet). Building on the classic Online Instance Classifier Refinement (OICR) framework, we propose to incorporate an contextual adaptation perception, within a multi-view learning framework, and integrate a pseudo-label filtering process. The contextual adaptation perception utilizes the surrounding environment information to enhance localization capabilities, guiding the model to prioritize target objects by referring to their spatially neighbouring pixels. Multi-view learning manufactures additional constraints from diverse perspectives, thereby revealing objects that might be overlooked due to the weak supervision in a single view. The pseudo-label filtering process eliminates inaccurate pseudo-labels by identifying reliable foregrounds to mitigate overlapping proposals during the label propagation. On challenging datasets NWPU VHR-10.v2 and DIOR, we achieve promising results with mAP of 62.3% and 28.2%, respectively, surpassing existing benchmarks.

关键词： weakly supervised object detection multi-view learning contextual adaptation remote sensing image

来源：评论

学校读者我要写书评

暂无评论

Misclassification in weakly supervised object detection

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2024年 33卷 3413-3427页

作者： Wu, Zhihao Xu, Yong Yang, Jian Li, Xuelong Harbin Inst Technol Shenzhen Sch Comp Sci & Technol Shenzhen 518055 Peoples R China Peng Cheng Lab Shenzhen 518055 Peoples R China Nanjing Univ Sci & Technol Dept Comp Sci & Engn Nanjing 210094 Peoples R China Northwestern Polytech Univ Sch Artificial Intelligence OPt & Elect iOPEN Xian 710072 Shaanxi Peoples R China

weakly supervised object detection (WSOD) aims to train detectors using only image-category labels. Current methods typically first generate dense class-agnostic proposals and then select objects based on the classification scores of these proposals. These methods mainly focus on selecting the proposal having high Intersection-over-Union with the true object location, while ignoring the problem of misclassification, which occurs when some proposals exhibit semantic similarities with objects from other categories due to viewing perspective and background interference. We observe that the positive class that is misclassified typically has the following two characteristics: 1) It is usually misclassified as one or a few specific negative classes, and the scores of these negative classes are high;2) Compared to other negative classes, the score of the positive class is relatively high. Based on these two characteristics, we propose misclassification correction (MCC) and misclassification tolerance (MCT) respectively. In MCC, we establish a misclassification memory bank to record and summarize the class-pairs with high frequencies of potential misclassifications in the early stage of training, that is, cases where the score of a negative class is significantly higher than that of the positive class. In the later stage of training, when such cases occur and correspond to the summarized class-pairs, we select the top-scoring negative class proposal as the positive training example. In MCT, we decrease the loss weights of misclassified classes in the later stage of training to avoid them dominating training and causing misclassification of objects from other classes that are semantically similar to them during inference. Extensive experiments on the PASCAL VOC and MS COCO demonstrate our method can alleviate the problem of misclassification and achieve the state-of-the-art results.

关键词： weakly supervised object detection misclassification semantics

来源：评论

学校读者我要写书评

暂无评论

weakly supervised object detection with Position Information of Convolution Neural Network 10

Weakly Supervised Object Detection with Position Information...

引用

IEEE 10th International Conference on Information, Communication and Networks (ICICN)

作者： Sun, Bo Yan, Huanqing He, Jun Yu, Lejun Zhang, Yinghui Beijing Normal Univ Sch Artif Intelligence Coll Educ Future Beijing Zhuhai Peoples R China Beijing Normal Univ Sch Artif Intelligence Beijing Peoples R China Beijing Normal Univ Coll Educ Future Zhuhai Peoples R China

ISBN: (纸本)9781665490825

weakly supervised object detection (WSOD) is a task that training object detector with only image-level label which typically means the category label. In this task, it is difficult to train a separate position regressor due to the lack of bounding box label. To solve this problem, some methods based on Multiple Instance Learning which establishes an instance (box) classifier by learning information from bag (image) are proposed by transforming the detection task into classification one. However, classifiers tend to find the most discriminative local region of an object, which leads to the problem of local optimization. In this paper, we utilize a regularization restriction method to obtain more global attention information by using the multi-layer features of convolutional neural network. Further, we use the feature enhanced by attention to locate more complete object. Meanwhile, the method is like a model plug-in which is easy to implement without additional computation. The result of the proposed method on PASCAL VOC2007 is better than baseline method, which has proven its effectiveness.

关键词： weakly supervised object detection multiple instance learning local optimization attention information

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：