检索结果-内蒙古大学图书馆

Lprr: Locality Preserving Robust Regression Based Sparse Feature Extraction

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Lai, Zhihui Zhu, Yufei Zhou, Jie Kong, Heng The College of Computer Science and Software Engineering Shenzhen University Guangdong Shenzhen518060 China The Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen University Guangdong Shenzhen518060 China The Department of Breast and Thyroid Surgery BaoAn Central Hospital of Shenzhen Guangdong Shenzhen518102 China

Jointly sparse projection learning attracts considerable attention due to its strong interpretability in feature extraction. To address the challenges related to weak discriminating representation in supervised feature extraction, we propose a more powerful regression framework. Based on the framework, we exhibit a new regression model called locality preserving robust regression (LPRR). In LPRR, we first construct an intrinsic graph to preserve the local geometric structure. Then, we ingeniously combine the reconstruction error minimization and projection variance maximization to explore the structured information of the data. Furthermore, the label information is utilized to explore the latent correlation structures among different classes and the L2,1-norm is applied to measure the regression and regularization terms, ensuring the joint sparsity of the projection matrix and enhancing the robustness of the model. An iterative algorithm is elaborately designed to achieve the optimal solution of LPRR, in which the subproblem of LPRR can be regarded as a general quadratic problem on the Stiefel manifold. The convergence and computational complexity of the proposed method LPRR are analyzed rigorously. In addition, extensive experiments demonstrate the competitive performance and some latent properties of LPRR. © 2023, The Authors. All rights reserved.

关键词： Feature extraction

Social Interpretable Tree for Pedestrian Trajectory Prediction

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Shi, Liushuai Wang, Le Long, Chengjiang Zhou, Sanping Zheng, Fang Zheng, Nanning Hua, Gang School of Software Engineering Xi'an Jiaotong University China Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University China JD Finance America Corporation United States Wormpex AI Research United States

Understanding the multiple socially-acceptable future behaviors is an essential task for many vision applications. In this paper, we propose a tree-based method, termed as Social Interpretable Tree (SIT), to address this multi-modal prediction task, where a hand-crafted tree is built depending on the prior information of observed trajectory to model multiple future trajectories. Specifically, a path in the tree from the root to leaf represents an individual possible future trajectory. SIT employs a coarse-to-fine optimization strategy, in which the tree is first built by high-order velocity to balance the complexity and coverage of the tree and then optimized greedily to encourage multimodality. Finally, a teacher-forcing refining operation is used to predict the final fine trajectory. Compared with prior methods which leverage implicit latent variables to represent possible future trajectories, the path in the tree can explicitly explain the rough moving behaviors (e.g., go straight and then turn right), and thus provides better interpretability. Despite the hand-crafted tree, the experimental results on ETH-UCY and Stanford Drone datasets demonstrate that our method is capable of matching or exceeding the performance of state-of-the-art methods. Interestingly, the experiments show that the raw built tree without training outperforms many prior deep neural network based approaches. Meanwhile, our method presents sufficient flexibility in long-term prediction and different best-of-K predictions. Code: https://***/lssiair/SIT. Copyright © 2022, The Authors. All rights reserved.

关键词： Trajectories

Hierarchical Decompositions of Stochastic Pursuit-Evasion Games

学校读者我要写书评

暂无评论

Hierarchical Decompositions of Stochastic Pursuit-Evasion Ga...

IEEE Conference on Decision and Control

作者： Yue Guan Mohammad Afshari Qifan Zhang Panagiotis Tsiotras Candidate With the School of Aerospace Engineering Georgia Institute of Technology Atlanta GA USA Postdoctoral Fellow With the Institute for Robotics and Intelligent Machines Georgia Institute of Technology Atlanta GA USA Currently a Software Engineer at Google Mountain View CA USA Georgia Tech School of Aerospace Engineering Georgia Institute of Technology Atlanta GA USA

ISBN: (数字)9781665467612

ISBN: (纸本)9781665467629

In this work we present a hierarchical framework for solving discrete stochastic pursuit-evasion games (PEGs) in large grid worlds. Given a partition of the grid world into superstates, the proposed approach creates a two-resolution decision-making process, which consists of a set of local PEGs at the original state level and an aggregated PEG at the superstate level. With much smaller state spaces, both the local games and the aggregated game can be easily solved to Nash equilibria. Through numerical simulations, we show that the proposed hierarchical framework significantly reduces the computation overhead, while still maintaining a satisfactory performance.

关键词： Decision making Games Numerical simulation

Delving into the Scale Variance Problem in Object Detection

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Chen, Junliang Zhao, Xiaodong Shen, Linlin Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence of Robotics of Society Shenzhen China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen 518060 China

Object detection has made substantial progress in the last decade, due to the capability of convolution in extracting local context of objects. However, the scales of objects are diverse and current convolution can only process single-scale input. The capability of traditional convolution with a fixed receptive field in dealing with such a scale variance problem, is thus limited. Multi-scale feature representation has been proven to be an effective way to mitigate the scale variance problem. Recent researches mainly adopt partial connection with certain scales, or aggregate features from all scales and focus on the global information across the scales. However, the information across spatial and depth dimensions is ignored. Inspired by this, we propose the multi-scale convolution (MSConv) to handle this problem. Taking into consideration scale, spatial and depth information at the same time, MSConv is able to process multi-scale input more comprehensively. MSConv is effective and computationally efficient, with only a small increase of computational cost. For most of the single-stage object detectors, replacing the traditional convolutions with MSConvs in the detection head can bring more than 2.5% improvement in AP (on COCO 2017 dataset), with only 3% increase of FLOPs. MSConv is also flexible and effective for two-stage object detectors. When extended to the mainstream two-stage object detectors, MSConv can bring up to 3.0% improvement in AP. Our best model under single-scale testing achieves 48.9% AP on COCO 2017 test-dev split, which surpasses many state-of-the-art methods. © 2022, CC BY.

关键词： Object detection

Selective Multi-Scale Learning for Object Detection

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Chen, Junliang Lu, Weizeng Shen, Linlin Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence of Robotics of Society Shenzhen China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen518060 China

Pyramidal networks are standard methods for multi-scale object detection. Current researches on feature pyramid networks usually adopt layer connections to collect features from certain levels of the feature hierarchy, and do not consider the significant differences among them. We propose a better architecture of feature pyramid networks, named selective multi-scale learning (SMSL), to address this issue. SMSL is efficient and general, which can be integrated in both single-stage and two-stage detectors to boost detection performance, with nearly no extra inference cost. RetinaNet combined with SMSL obtains 1.8% improvement in AP (from 39.1% to 40.9%) on COCO dataset. When integrated with SMSL, two-stage detectors can get around 1.0% improvement in AP. © 2022, CC BY.

关键词： Object detection

Spatio-temporal Collaborative Convolution for Video Action Recognition

学校读者我要写书评

暂无评论

Spatio-temporal Collaborative Convolution for Video Action R...

2020 IEEE International Conference on Artificial Intelligence and Computer Applications, ICAICA 2020

作者： Li, Xu Wen, Liqiang Wang, Jinjun Zeng, Ming School of Software Engineering Xi'An Jiaotong University Xi'an China School of Software and Microelectronics Peking University Beijing China Institute of Artificial Intelligence and Robotics Xi'An Jiaotong University Xi'an China

ISBN: (纸本)9781728170046

Although video action recognition has achieved great progress in recent years, it is still a challenging task due to the huge computational complexity. Designing a lightweight network is a feasible solution, but it may reduce the spatio-temporal information modeling capability. In this paper, we propose a novel novel spatio-temporal collaborative convolution (denote as 'STC-Conv'), which can efficiently encode spatio-temporal information. STC-Conv collaboratively learn spatial and temporal feature in one convolution filter kernel. In short, temporal convolution and spatial convolution are integrated in the one STC convolution kernel, which can effectively reduce the model complexity and improve the computational efficiency. STC-Conv is a universal convolution, which can be applied to the existing 2D CNNs, such as ResNet, DenseNet. The experimental results on the temporal-related dataset Something Something V1 prove the superiority of our method. Noticeably, STC-Conv enjoys more excellent performance than 3D CNNs at even lower computation cost than standard 2D CNNs. © 2020 IEEE.

关键词： Convolution

Unsupervised Video Anomaly Detection with Self-Attention Based Feature Aggregating

学校读者我要写书评

暂无评论

Unsupervised Video Anomaly Detection with Self-Attention Bas...

International Conference on Intelligent Transportation

作者： Zhenhao Ye Yanlong Li Zhichao Cui Yuehu Liu Li Li Le Wang Chi Zhang National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an Shaanxi China School of Software Engineering Xi'an Jiaotong University Xi'an Shaanxi China School of Electronics and Control Engineering Chang'an University Department of Automation BNRist Tsinghua University Beijing China

Anomaly detection in surveillance videos is a crucial and challenging task in the intelligent transportation systems. Previous methods utilize a memory module to store prototypical feature embeddings as normal patterns learned from normal training data. However, due to the complexity of real-world scenarios, it is difficult to choose an appropriate size of memory module which can not only memorize normal patterns comprehensively, but also capable of dealing with unseen normal scenarios. To tackle this problem, we learn normal video patterns by constructing and exploring correlations between visual semantics. In the training stage, we act the self-attention map between embeddings as a description of information association between different visual semantics. A self-attention based feature aggregating module is designed to regenerate a feature map through aggregating embeddings with similar information based on the attention map. By decoding the generated feature map instead of the original one to predict the future frame of the input video clip, the model learns to build strong information association between normal visual semantics. Moreover, we observe that abnormal embeddings hardly build strong association with others. Thus, we design a feature-level anomaly criterion referred as prior deviation to increase the difference between attention maps generated by normal and abnormal frames. In the inferring stage, the proposed prior deviation jointly detects anomalies with pixel-level frame prediction error. Experiment results and ablation studies on mainstream benchmarks demonstrate the effectiveness of our design.

关键词：

MCTSteg: A Monte Carlo tree search-based reinforcement learning framework for universal non-additive steganography

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Mo, Xianbo Tan, Shunquan Li, Bin Huang, Jiwu College of Computer Science and Software Engineering Shenzhen University College of Information Engineering Shenzhen University Shenzhen Institute of Artificial Intelligence and Robotics for Society China

Recent research has shown that non-additive image steganographic frameworks effectively improve security performance through adjusting distortion distribution. However, as far as we know, all of the existing non-additive proposals are based on handcrafted policies, and can only be applied to a specific image domain, which heavily prevent non-additive steganography from releasing its full potentiality. In this paper, we propose an automatic non-additive steganographic distortion learning framework called MCTSteg to remove the above restrictions. Guided by the reinforcement learning paradigm, we combine Monte Carlo Tree Search (MCTS) and steganalyzer-based environmental model to build MCTSteg. MCTS makes sequential decisions to adjust distortion distribution without human intervention. Our proposed environmental model is used to obtain feedbacks from each decision. Due to its self-learning characteristic and domain-independent reward function, MCTSteg has become the first reported universal non-additive steganographic framework which can work in both spatial and JPEG domains. Extensive experimental results show that MCTSteg can effectively withstand the detection of both hand-crafted feature-based and deep-learning-based steganalyzers. In both spatial and JPEG domains, the security performance of MCTSteg steadily outperforms the state of the art by a clear margin under different scenarios. Copyright © 2021, The Authors. All rights reserved.

关键词： Reinforcement learning

A survey: Deep learning for hyperspectral image classification with few labeled samples

学校读者我要写书评

暂无评论

Neurocomputing 2021年 448卷 179-204页

作者： Sen Jia Shuguo Jiang Zhijie Lin Nanying Li Meng Xu Shiqi Yu College of Computer Science and Software Engineering Shenzhen University China SZU Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society China Department of Computer Science and Engineering Southern University of Science and Technology China

With the rapid development of deep learning technology and improvement in computing capability, deep learning has been widely used in the field of hyperspectral image (HSI) classification. In general, deep learning models often contain many trainable parameters and require a massive number of labeled samples to achieve optimal performance. However, in regard to HSI classification, a large number of labeled samples is generally difficult to acquire due to the difficulty and time-consuming nature of manual labeling. Therefore, many research works focus on building a deep learning model for HSI classification with few labeled samples. In this article, we concentrate on this topic and provide a systematic review of the relevant literature. Specifically, the contributions of this paper are twofold. First, the research progress of related methods is categorized according to the learning paradigm, including transfer learning, active learning and few-shot learning. Second, a number of experiments with various state-of-the-art approaches has been carried out, and the results are summarized to reveal the potential research directions. More importantly, it is notable that although there is a vast gap between deep learning models (that usually need sufficient labeled samples) and the HSI scenario with few labeled samples, the issues of small-sample sets can be well characterized by fusion of deep learning methods and related techniques, such as transfer learning and a lightweight model. For reproducibility, the source codes of the methods assessed in the paper can be found at https://***/ShuGuoJ/***.

关键词： Hyperspectral image classification Deep learning Transfer learning Few-shot learning