检索结果-内蒙古大学图书馆

Real-time unsupervised video object detection on the edge

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE 2025年 167卷

作者： Ruiz-Barroso, Paula Castro, Francisco M. Guil, Nicolas Univ Malaga Dept Comp Architecture Malaga Spain

object detection in video is an essential computer vision task. Consequently, many efforts have been devoted to developing precise and fast deep-learning models for this task. These models are commonly deployed on discrete and powerful GPU devices to meet both frame rate performance and detection accuracy requirements. Furthermore, model training is usually performed in a strongly supervised way so that samples must be previously labelled by humans using a slow and costly process. In this paper, we develop a real-time implementation for unsupervised object detection in video employing a low-power device. We improve typical approaches for object detection using information supplied by optical flow to detect moving objects. Besides, we use an unsupervised clustering algorithm to group similar detections that avoid manual object labelling. Finally, we propose a methodology to optimize the deployment of our resulting framework on an embedded heterogeneous platform. Thus, we illustrate how all the computational resources of a Jetson AGX Xavier (CPU, GPU, and DLAs) can be used to fulfil frame rate, accuracy, and energy consumption requirements. Three different data representations (FP32, FP16 and INT8) are studied for the pipeline networks in order to evaluate the impact of all of them in our pipeline. Obtained results show that our proposed optimizations can improve up to 23.6x energy consumption and 32.2x execution time with respect to the non-optimized pipeline without penalizing the original mAP (59.44). This computational complexity reduction is achieved through knowledge distillation, using FP16 data precision, and deploying concurrent tasks indifferent computing units.

关键词： Embedded system unsupervised object detection Model quantization Knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

Semantic-aware representations for unsupervised Camouflaged object detection

引用

JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION 2025年 107卷

作者： Lu, Zelin Zhao, Xing Xie, Liang Liang, Haoran Liang, Ronghua Zhejiang Univ Technol Coll Comp Sci & Technol 288 Liuhe Rd Hangzhou 310023 Peoples R China

unsupervised image segmentation algorithms face challenges due to the lack of human annotations. They typically employ representations derived from self-supervised models to generate pseudo-labels for supervising model training. Using this strategy, the model's performance largely depends on the quality of the generated pseudo-labels. In this study, we design an unsupervised framework to perform COD (Camouflaged object detection) without the need for generating pseudo-labels. Specifically, we utilize semantic-aware representations, trained in a self-supervised manner on large-scale unlabeled datasets, to guide the training process. These representations not only capturing rich contextual semantic information but also assist in refining the blurred boundaries of camouflaged objects. Furthermore, we design a framework that integrates these semantic- aware representations with task-specific features, enabling the model to perform the UCOD (unsupervised Camouflaged object detection) task with enhanced contextual understanding. Moreover, we introduce an innovative multi-scale token loss function, which maintain the structural integrity of objects at various scales in the model's predictions through mutual supervision between different features and scales. Extensive experimental validation demonstrates that our model significantly enhances the performance of UCOD, closely approaching the capabilities of state-of-the-art weakly-supervised COD models.

关键词： Semantic-aware segmentation unsupervised object detection Camouflaged object detection

来源：评论

学校读者我要写书评

暂无评论

Novel patch selection based on object detection in HMAX for natural image classification

引用

SIGNAL IMAGE AND VIDEO PROCESSING 2022年第4期16卷 1101-1108页

作者： Akbarpour, Mohammadesmaeil Mandal, Mrinal Kamangar, M. Hashemi Shomal Univ Elect Engn Amol Iran Univ Alberta Elect & Comp Engn Edmonton AB Canada

The human visual system (HVS) can effectively recognize objects in complex natural scenes with high speed and accuracy. Many models have been proposed based on HVS among which HMAX is one of the superior models. In HMAX, the random extraction of a large volume of training samples, called patches, has two drawbacks. First, patches from background, in addition to high computational cost, can produce wrong output. Second, patches with low information from objects may provide poor performance. In this paper, an optimum method, with two steps, is proposed to select patches with high discriminative information. First, a pool of patches is extracted from objects based on an unsupervised object detection method. Second, patches with high discriminative information were selected from the pool based on patch ranking. Further, complement of optimum patch for each class is considered as a new patch for other classes to increase the recognition rate. Experimental results with Caltech5, Caltech101 and Graz-01 databases show that the proposed model provides a significant performance improvement over the HMAX and other state-of-the-art models, in terms of speed, sensitivity, specificity and classification accuracy.

关键词： Classification unsupervised object detection Human visual system HMAX Optimum patch

来源：评论

学校读者我要写书评

暂无评论

CroMoDa: unsupervised Oriented SAR Ship detection via Cross-Modality Distribution Alignment

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2024年 17卷 11899-11914页

作者： Chen, Xi Wang, Zhirui Wang, Wenhao Xie, Xinyi Kang, Jian Fernandez-Beltran, Ruben Soochow Univ Sch Elect & Informat Engn Suzhou 215006 Peoples R China Chinese Acad Sci Aerosp Informat Res Inst Beijing 100190 Peoples R China Chinese Acad Sci Inst Elect Key Lab Network Informat Syst Technol Beijing 100190 Peoples R China Zhejiang Univ Technol Coll Informat Engn Hangzhou 310023 Peoples R China Univ Murcia Dept Comp Sci & Syst Murcia 30100 Spain

Most state-of-the-art synthetic aperture radar (SAR) ship detection methods based on deep learning require large amounts of labeled data for network training. However, the annotation process requires significant manpower and resources especially for SAR images, since relevant background knowledge should be necessary for the annotators. Considering the available optical imagery datasets with labels, we propose an unsupervised oriented SAR ship detection method based on cross-modality distribution alignment, termed as CroMoDa. It consists of four components: 1) image-level feature alignment;2) low-level feature despeckling;3) cross-modality pseudo-label self-training;and 4) cross-modality object alignment. By aligning the multilevel feature distributions, modality-invariant features across the two imagery modalities can be learned. Considering speckle noise and other interferences in SAR images, the proposed loss term conducted on low-level features can enhance the object information to improve the detection accuracy. Moreover, the proposed pseudo-label self-training can better generate oriented SAR ship annotations than the other methods, which facilitates learning more discriminative features for SAR ship instances. With two optical-SAR dataset configurations, the proposed method is effectively evaluated by comparing to the other state of the arts, which demonstrates the great potential for SAR ship detection in real applications.

关键词： Marine vehicles Radar polarimetry Optical imaging Adaptive optics Synthetic aperture radar Feature extraction Training Cross modality synthetic aperture radar (SAR) synthetic aperture radar ship detection transfer learning unsupervised object detection

来源：评论

学校读者我要写书评

暂无评论

Automatic detection of geospatial objects using multiple hierarchical segmentations

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2008年第7期46卷 2097-2111页

作者： Akcay, H. Goekhan Aksoy, Selim Bilkent Univ Dept Comp Engn TR-06800 Ankara Turkey

The object-based analysis of remotely sensed imagery provides valuable spatial and structural information that is complementary to pixel-based spectral information in classification. In this paper, we present novel methods for automatic object detection in high-resolution images by combining spectral information with structural information exploited by using image segmentation. The proposed segmentation algorithm uses morphological operations applied to individual spectral bands using structuring elements in increasing sizes. These operations produce a set of connected components forming a hierarchy of segments for each band. A generic algorithm is designed to select meaningful segments that maximize a measure consisting of spectral homogeneity and neighborhood. connectivity. Given the observation that different structures appear more clearly at different scales in different spectral bands, we describe a new algorithm for unsupervised grouping of candidate segments belonging to multiple hierarchical segmentations to find coherent sets of segments that correspond to actual objects. The segments are modeled by using their spectral and textural content, and the grouping problem is solved by using the probabilistic latent semantic analysis algorithm that builds object models by learning the object-conditional probability distributions. The automatic labeling of a segment is done by computing the similarity of its feature distribution to the distribution of the learned object models using the Kullback-Leibler divergence. The performances of the unsupervised segmentation and object detection algorithms are evaluated qualitatively and quantitatively using three different data sets with comparative experiments, and the results show that the proposed methods are able to automatically detect, group, and label segments belonging to the same object classes.

关键词： hierarchical segmentation image segmentation mathematical morphology object-based analysis unsupervised object detection

来源：评论

学校读者我要写书评

暂无评论

Subfeature Ensemble-Based Hyperspectral Anomaly detection Algorithm

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2022年 15卷 5943-5952页

作者： Wang, Shuo Feng, Wei Quan, Yinghui Bao, Wenxing Dauphin, Gabriel Gao, Lianru Zhong, Xian Xing, Mengdao Xidian Univ Dept Remote Sensing Sci & Technol Sch Elect Engn Xian 710071 Peoples R China North Minzu Univ Sch Comp Sci & Engn Yinchuan 750021 Ningxia Peoples R China Univ Paris XIII Lab Informat Proc & Transmiss L2TI Inst Galilee F-93430 Villetaneuse France Chinese Acad Sci Key Lab Digital Earth Sci Aerosp Informat Res Inst Beijing 100094 Peoples R China Xidian Univ Acad Adv Interdisciplinary Res Xian 710071 Peoples R China

Hyperspectral images (HSIs) have always played an important role in remote sensing applications. Anomaly detection has become a hot spot in HSI processing in recent years. The popular detecting method is to accurately segment anomalies from the background. Informative bands are very important for the accuracy improvement of the detection technology. However, most of the abnormal targets segmentation methods focus on the usage of all the spectral features, thus are easily affected by redundant bands or feature noise. A hyperspectral anomaly detection algorithm based on subfeature ensemble is proposed in this article. The proposed method consists of the following steps. First, the bands of the original HSI are normalized and randomly divided into several subfeature sets according to different proportions. Second, six methods including the prior-based tensor approximation algorithm (PTA), Reed-Xiaoli method, a low-rank and sparse representation method, a low-rank and sparse matrix decomposition-based Mahalanobis distance method, the graph and total variation regularized low-rank representation-based method, and a method based on tensor principal component analysis are applied to detect anomalies on the original HSI, and the method with the best performance is used to obtain an enhanced feature set. Then, the enhanced features and the subfeatures are ensembled iteratively to construct a new dataset. Finally, the PTA method is operated on the dataset with ensemble features to get the final abnormal target results. Six hyperspectral datasets are used in the experiment. Seven methods are employed as comparisons. The results are analyzed from both qualitative and quantitative perspectives. Extensive experimental results illustrate that the proposed method performs best on all datasets.

关键词： Anomaly detection Tensors Feature extraction Sparse matrices Hyperspectral imaging object detection Matrix decomposition Anomaly detection hyperspectral images (HSIs) feature ensemble unsupervised object detection remote sensing

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Vision Architecture for AGI Systems 16th

An Adaptive Vision Architecture for AGI Systems

引用

16th International Conference on Artificial General Intelligence (AGI)

作者： Wunsche, Robert Stockholm Univ Dept Psychol Stockholm Sweden

ISBN: (纸本)9783031334689;9783031334696

This paper presents an unsupervised object detection system which can offline-learn generic visual features via Siamese neural network, yet is able to learn new object classes at run-time with a prototype learning approach applied on the latent representations. The operating requirements of this system feature bounded processing time per frame, while dealing with a fixed amount of available memory. This system works under the Assumption of Insufficient Knowledge and Resource and is hence operating in real-time and open to new information which can arrive at any time, as systems such as NARS and AERA ideally also demand for perception.

关键词： unsupervised object detection Vision System Real-time vision Prototype formation Siamese Neural Networks

来源：评论

学校读者我要写书评

暂无评论

OCIE: Augmenting model interpretability via Deconfounded Explanation-Guided Learning

引用

KNOWLEDGE-BASED SYSTEMS 2024年 302卷

作者： Dong, Liang Chen, Leiyang Zheng, Chengliang Fu, Zhongwang Zukaib, Umer Cui, Xiaohui Shen, Zhidong Wuhan Univ Sch Cyber Sci & Engn Wuhan 430000 Peoples R China Minist Educ Key Lab Aerosp Informat Secur & Trusted Comp Wuhan 430000 Peoples R China

Deep neural networks (DNNs) often encounter significant challenges related to opacity, inherent biases, and shortcut learning, which undermine their practical reliability. In this study, we address these issues by constructing a causal graph to model the unbiased learning process of DNNs. This model reveals that recurrent background information in training samples acts as a confounder, leading to spurious correlations between model inputs and outputs, causing biased predictions. To mitigate these problems and promote unbiased feature learning, we propose the object-guided Consistency Interpretation Enhancement (OCIE) methodology. OCIE enhances DNN interpretability by integrating explicit objects and explanations into the model's learning process. Initially, OCIE employs a graph-based algorithm to identify explicit objects within self-supervised vision transformer-learned features. Subsequently, it constructs class prototypes to eliminate invalid detected objects. Finally, OCIE aligns explanations with explicit objects, directing the model's attention towards the most distinctive classification features rather than irrelevant backgrounds. Extensive experiments on different image classification datasets, including general (ImageNet), fine-grained (Stanford Cars and CUB-200), and medical (HAM) datasets, using two prevailing network architectures, demonstrate that OCIE significantly enhances explanation consistency across all datasets. Furthermore, OCIE proves particularly advantageous for fine-grained classification, especially in few-shot scenarios, by improving both interpretability and classification performance. Additionally, our findings highlight the impact of centralized explanations on the sufficiency of model decisions, suggesting that focusing explanations on explicit objects improves the reliability of DNN predictions. Our code is available at: https://***/DLAIResearch/OCIE.

关键词： Explanation-guided learning Causal inference Explainable Artificial Intelligence unsupervised object detection Image classification

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：