检索结果-内蒙古大学图书馆

Entropy-based Active Learning for Object Detection with Progressive Diversity Constraint

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wu, Jiaxi Chen, Jiaxin Huang, Di State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China

Active learning is a promising alternative to alleviate the issue of high annotation cost in the computer vision tasks by consciously selecting more informative samples to lab.l. Active learning for object detection is more challenging and existing efforts on it are relatively rare. In this paper, we propose a novel hybrid approach to address this problem, where the instance-level uncertainty and diversity are jointly considered in a bottom-up manner. To balance the computational complexity, the proposed approach is designed as a two-stage procedure. At the first stage, an Entropy-based Non-Maximum Suppression (ENMS) is presented to estimate the uncertainty of every image, which performs NMS according to the entropy in the feature space to remove predictions with redundant information gains. At the second stage, a diverse prototype (DivProto) strategy is explored to ensure the diversity across images by progressively converting it into the intra-class and inter-class diversities of the entropy-based class-specific prototypes. Extensive experiments are conducted on MS COCO and Pascal VOC, and the proposed approach achieves state of the art results and significantly outperforms the other counterparts, highlighting its superiority. Copyright © 2022, The Authors. All rights reserved.

关键词： Object recognition

INITNO: Boosting Text-to-Image Diffusion Models via Initial Noise Optimization

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Guo, Xiefan Liu, Jinlin Cui, Miaomiao Li, Jiankai Yang, Hongyu Huang, Di State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China Institute of Artificial Intelligence Beihang University Beijing China Shanghai Artificial Intelligence Laboratory Shanghai China

Recent strides in the development of diffusion models, exemplified by advancements such as Stable Diffusion, have underscored their remarkable prowess in generating visually compelling images. However, the imperative of achieving a seamless alignment between the generated image and the provided prompt persists as a formidable challenge. This paper traces the root of these difficulties to invalid initial noise, and proposes a solution in the form of Initial Noise Optimization (INITNO), a paradigm that refines this noise. Considering text prompts, not all random noises are effective in synthesizing semantically-faithful images. We design the cross-attention response score and the self-attention conflict score to evaluate the initial noise, bifurcating the initial latent space into valid and invalid sectors. A strategically crafted noise optimization pipeline is developed to guide the initial noise towards valid regions. Our method, validated through rigorous experimentation, shows a commendable proficiency in generating images in strict accordance with text prompts. Our code is availab.e at https://***/xiefan-guo/initno. Copyright © 2024, The Authors. All rights reserved.

关键词： Diffusion

Initno: Boosting Text-to-Image Diffusion Models via Initial Noise Optimization

学校读者我要写书评

暂无评论

Initno: Boosting Text-to-Image Diffusion Models via Initial ...

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Xiefan Guo Jinlin Liu Miaomiao Cui Jiankai Li Hongyu Yang Di Huang State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China Institute of Artificial Intelligence Beihang University Beijing China Shanghai Artificial Intelligence Laboratory Shanghai China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Recent strides in the development of diffusion models, ex-emplified by advancements such as Stable Diffusion, have underscored their remarkable prowess in generating visu-ally compelling images. However, the imperative of achieving a seamless alignment between the generated image and the provided prompt persists as a formidable challenge. This paper traces the root of these difficulties to invalid initial noise, and proposes a solution in the form of Initial Noise Optimization (INITNO), a paradigm that refines this noise. Considering text prompts, not all random noises are effective in synthesizing semantically-faithful images. We design the cross-attention response score and the selfattention conflict score to evaluate the initial noise, bifurcating the initial latent space into valid and invalid sectors. A strategically crafted noise optimization pipeline is developed to guide the initial noise towards valid regions. Our method, validated through rigorous experimentation, shows a commendable proficiency in generating images in strict accordance with text prompts. Our code is availab.e at https://***/xiefan-guo/initno.

关键词： computer vision Navigation Instruments Noise Pipelines Text to image Aerospace electronics

Editorial: Efficient deep neural network for intelligent robot system: Focusing on visual signal processing

学校读者我要写书评

暂无评论

Frontiers in Neurorobotics 2023年 17卷 1191655页

作者： Bai, Xiao Ning, Xin Donta, Praveen Kumar Li, Weijun School of Computer Science and Engineering State Key Laboratory of Software Development Environment Jiangxi Research Institute Beihang University Beijing China Institute of Semiconductors Chinese Academy of Sciences Beijing China Vienna Austria

can be effectively applied to facial expression recognition. Lu et al. present a facial image inpainting method using a multistage GAN and the global attention mechanism. The proposed method can effectively restore incomplete facial images by enhancing feature mining and semantic expression, using skip connections, encoder-decoder structure, and a local refinement network. Comparative experiments demonstrate that the proposed method generates realistic inpainting results with high PSNR and SSIM, indicating the model's performance and efficiency. Lin et al. address the shortage of boxing coaches in Chinese campuses by proposing a novel solution that employs human pose estimation technology to train interns. Specifically, they develop a model transfer technique that utilizes channel patching to enhance the accuracy of pose key points by an average of 1-20% and 3D accuracies by 0.3-0.5% compared to 2D baselines. The proposed method is not only practical but also effective for boxing pose *** contributions focus on the implementation of lightweight models to address other signals. Zheng presents a writing feature abstraction process based on ON-LSTM and attention mechanism for sentiment analysis, addressing the problem of ignoring syntactic and tag semantics information in emotional text feature extraction. The study shows the high application potential of deep learning models for dynamic user sentiment analysis. Wang and Chen investigate teachers' acceptance of robotics education and its relationship to the effectiveness and sustainability of robotics education using the UTAUT model and deep learning algorithms. The study also found that deep learning models such as mDAE and AmDAE reduced training time compared to existing noise-reducing autoencoder models. Teng et al. address the lack of technical and algorithmic support in music therapy for cancer patients and design a neural network robotic system based on breast cancer patients to analyze the ef

关键词： deep neural network intelligent robot system image processing pattern recognition visual navigation

Multi-View Incongruity Learning for Multimodal Sarcasm Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Guo, Diandian Cao, Cong Yuan, Fangfang Liu, Yanbing Zeng, Guangjie Yu, Xiaoyan Peng, Hao Yu, Philip S. Institute of Information Engineering Chinese Academy of Sciences China School of Cyber Security University of Chinese Academy of Sciences China State Key Laboratory of Software Development Environment Beihang University China School of Computer Science and Technology Beijing Institute of Technology China Department of Computer Science University of Illinois Chicago United States

Multimodal sarcasm detection (MSD) is essential for various downstream tasks. Existing MSD methods tend to rely on spurious correlations. These methods often mistakenly prioritize non-essential features yet still make correct predictions, demonstrating poor generalizability beyond training environments. Regarding this phenomenon, this paper undertakes several initiatives. Firstly, we identify two primary causes that lead to the reliance of spurious correlations. Secondly, we address these challenges by proposing a novel method that integrate Multimodal Incongruities via Contrastive Learning (MICL) for multimodal sarcasm detection. Specifically, we first leverage incongruity to drive multi-view learning from three views: token-patch, entity-object, and sentiment. Then, we introduce extensive data augmentation to mitigate the biased learning of the textual modality. Additionally, we construct a test set, SPMSD, which consists potential spurious correlations to evaluate the the model’s generalizability. Experimental results demonstrate the superiority of MICL on benchmark datasets, along with the analyses showcasing MICL’s advancement in mitigating the effect of spurious correlation. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

Student-Teacher Feature Pyramid Matching for Anomaly Detection 32

学校读者我要写书评

暂无评论

Student-Teacher Feature Pyramid Matching for Anomaly Detecti...

32nd British Machine Vision Conference, BMVC 2021

作者： Wang, Guodong Han, Shumin Ding, Errui Huang, Di State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China Department of Computer Vision Technology Baidu Inc. Beijing China

Anomaly detection is a challenging task and usually formulated as an one-class learning problem for the unexpectedness of anomalies. This paper proposes a simple yet powerful approach to this issue, which is implemented in the student-teacher framework for its advantages but substantially extends it in terms of both accuracy and efficiency. Given a strong model pre-trained on image classification as the teacher, we distill the knowledge into a single student network with the identical architecture to learn the distribution of anomaly-free images and this one-step transfer preserves the crucial clues as much as possible. Moreover, we integrate the multi-scale feature matching strategy into the framework, and this hierarchical feature matching enables the student network to receive a mixture of multi-level knowledge from the feature pyramid under better supervision, thus allowing to detect anomalies of various sizes. The difference between feature pyramids generated by the two networks serves as a scoring function indicating the probability of anomaly occurring. Due to such operations, our approach achieves accurate and fast pixel-level anomaly detection. Very competitive results are delivered on the MVTec anomaly detection dataset, superior to the state of the art ones. © 2021. The copyright of this document resides with its authors.

关键词： Anomaly detection

Understanding Heterophily for Graph Neural Networks

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Junfu Guo, Yuanfang Yang, Liang Wang, Yunhong State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China Shen Yuan Honors College Beihang University Beijing China School of Artificial Intelligence Hebei University of Technology Tianjin China

Graphs with heterophily have been regarded as challenging scenarios for Graph Neural Networks (GNNs), where nodes are connected with dissimilar neighbors through various patterns. In this paper, we present theoretical understandings of heterophily for GNNs by incorporating the graph convolution (GC) operations into fully connected networks via the proposed Heterophilous Stochastic Block Models (HSBM), a general random graph model that can accommodate diverse heterophily patterns. Our theoretical investigation comprehensively analyze the impact of heterophily from three critical aspects. Firstly, for the impact of different heterophily patterns, we show that the separability gains are determined by two factors, i.e., the Euclidean distance of the neighborhood distributions and pE [deg], where E [deg] is the averaged node degree. Secondly, we show that the neighborhood inconsistency has a detrimental impact on separability, which is similar to degrading E [deg] by a specific factor. Finally, for the impact of stacking multiple layers, we show that the separability gains are determined by the normalized distance of the lpowered neighborhood distributions, indicating that nodes still possess separability in various regimes, even when over-smoothing occurs. Extensive experiments on both synthetic and real-world data verify the effectiveness of our theory. Copyright © 2024, The Authors. All rights reserved.

关键词： Graph neural networks

Boundary Guided Context Aggregation for Semantic Segmentation 32

学校读者我要写书评

暂无评论

Boundary Guided Context Aggregation for Semantic Segmentatio...

32nd British Machine Vision Conference, BMVC 2021

作者： Ma, Haoxiang Yang, Hongyu Huang, Di State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China Institute of Artificial Intelligence Beihang University Beijing China

The recent studies on semantic segmentation are starting to notice the significance of the boundary information, where most approaches see boundaries as the supplement of semantic details. However, simply combing boundaries and the mainstream features cannot ensure a holistic improvement of semantics modeling. In contrast to the previous studies, we exploit boundary as a significant guidance for context aggregation to promote the overall semantic understanding of an image. To this end, we propose a Boundary guided Context Aggregation Network (BCANet), where a Multi-Scale Boundary extractor (MSB) borrowing the backbone features at multiple scales is specifically designed for accurate boundary detection. Based on which, a Boundary guided Context Aggregation module (BCA) improved from Non-local network is further proposed to capture long-range dependencies between the pixels in the boundary regions and the ones inside the objects. By aggregating the context information along the boundaries, the inner pixels of the same category achieve mutual gains and therefore the intra-class consistency is enhanced. We conduct extensive experiments on the Cityscapes and ADE20K databases, and comparable results are achieved with the state-of-the-art methods, clearly demonstrating the effectiveness of the proposed one. Our code is availab.e at https://***/mahaoxiang822/Boundary-Guided-Context-Aggregation. © 2021. The copyright of this document resides with its authors.

关键词： Semantic Segmentation

Boosting performance of virtualized desktop infrastructure with physical GPU and SPICE

学校读者我要写书评

暂无评论

science China(Information sciences) 2020年第7期63卷 236-238页

作者： Shupan LI Limin XIAO Chungang SHI Liequan CHE Changyou ZHANG Yuanzhang LI State Key Laboratory of Software Development Environment Beihang University School of Computer Science and Engineering Beihang University Beijing Jinghang Computation and Communication Research Institute Institute of Software Chinese Academy of Sciences School of Computer Science and Technology Beijing Institute of Technology

Dear editor,The virtual desktop infrastructure (VDI) is widely used by enterprises, as it can be easily deployed,has a reduced management cost, improves resource utilization, and can be accessed at any location by the network. Currently, there is a demand for the VDIs to support applications, such as playing 3D games, playing HD videos, and using 3D design-

关键词： visualized desktop infrastructure SPICE physical GPU The virtual machine (VM) virtual machine monitor (VMM)