检索结果-内蒙古大学图书馆

A Holistically-Guided decoder for Deep Representation Learning With Applications to Semantic Segmentation and Object Detection

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2023年第10期45卷 11390-11406页

作者： Liu, Jianbo He, Junjun Zheng, Yuanjie Yi, Shuai Wang, Xiaogang Li, Hongsheng Chinese Univ Hong Kong Dept Elect Engn Hong Kong Peoples R China SenseTime Res Shanghai 200233 Peoples R China Shandong Normal Univ Sch Informat Sci & Engn Jinan 250014 Shandong Peoples R China

Both high-level and high-resolution feature representations are of great importance in various visual understanding tasks. To acquire high-resolution feature maps with high-level semantic information, one common strategy is to adopt dilated convolutions in the backbone networks to extract high-resolution feature maps, such as the dilatedFCN-based methods for semantic segmentation. However, due to many convolution operations are conducted on the high-resolution feature maps, such methods have large computational complexity and memory consumption. To balance the performance and efficiency, there also exist encoder-decoder structures that gradually recover the spatial information by combining multi-level feature maps from a feature encoder, such as the FPN architecture for object detection and the U-Net for semantic segmentation. Although being more efficient, the performances of existing encoder-decoder methods for semantic segmentation are far from comparable with the dilatedFCN-based methods. In this paper, we propose one novel holistically-guided decoder which is introduced to obtain the high-resolution semantic-rich feature maps via the multi-scale features from the encoder. The decoding is achieved via novel holistic codeword generation and codeword assembly operations, which take advantages of both the high-level and low-level features from the encoder features. With the proposed holistically-guided decoder, we implement the EfficientFCN architecture for semantic segmentation and HGD-FPN for object detection and instance segmentation. The EfficientFCN achieves comparable or even better performance than state-of-the-art methods with only 1/3 of their computational costs for semantic segmentation on PASCAL Context, PASCAL VOC, ADE20K datasets. Meanwhile, the proposed HGD-FPN achieves > 2% higher mean Average Precision (mAP) when integrated into several object detection frameworks with ResNet-50 encoding backbones.

关键词： Semantic segmentation object detection encoder-decoder dilated convolution holistic features feature pyramids

来源：评论

学校读者我要写书评

暂无评论

A Sentence Retrieval Generation Network Guided Video Captioning

引用

Computers, Materials & Continua 2023年第6期75卷 5675-5696页

作者： Ou Ye Mimi Wang Zhenhua Yu Yan Fu Shun Yi Jun Deng College of Computer Science and Technology Xi’an University of Science and TechnologyXi’an710054China College of Safety and Engineering Xi’an University of Science and TechnologyXi’an710054China

Currently,the video captioning models based on an encoder-decoder mainly rely on a single video input *** contents of video captioning are limited since few studies employed external corpus information to guide the generation of video captioning,which is not conducive to the accurate descrip-tion and understanding of video *** address this issue,a novel video captioning method guided by a sentence retrieval generation network(ED-SRG)is proposed in this ***,a ResNeXt network model,an efficient convolutional network for online video understanding(ECO)model,and a long short-term memory(LSTM)network model are integrated to construct an encoder-decoder,which is utilized to extract the 2D features,3D features,and object features of video data *** features are decoded to generate textual sentences that conform to video content for sentence ***,a sentence-transformer network model is employed to retrieve different sentences in an external corpus that are semantically similar to the above textual *** candidate sentences are screened out through similarity ***,a novel GPT-2 network model is constructed based on GPT-2 network *** model introduces a designed random selector to randomly select predicted words with a high probability in the corpus,which is used to guide and generate textual sentences that are more in line with human natural language *** proposed method in this paper is compared with several existing works by *** results show that the indicators BLEU-4,CIDEr,ROUGE_L,and METEOR are improved by 3.1%,1.3%,0.3%,and 1.5%on a public dataset MSVD and 1.3%,0.5%,0.2%,1.9%on a public dataset MSR-VTT *** can be seen that the proposed method in this paper can generate video captioning with richer semantics than several state-of-the-art approaches.

关键词： Video captioning encoder-decoder sentence retrieval external corpus RS GPT-2 network model

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Graph Pattern Understanding for Zero-Shot Video Object Segmentation

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2023年 32卷 5909-5920页

作者： Pei, Gensheng Shen, Fumin Yao, Yazhou Chen, Tao Hua, Xian-Sheng Shen, Heng-Tao Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing 210094 Peoples R China Univ Elect Sci & Technol China Sch Comp Sci & Engn Chengdu 610056 Peoples R China DAMO Acad Alibaba Grp Hangzhou 100080 Peoples R China

The optical flow guidance strategy is ideal for obtaining motion information of objects in the video. It is widely utilized in video segmentation tasks. However, existing optical flow-based methods have a significant dependency on optical flow, which results in poor performance when the optical flow estimation fails for a particular scene. The temporal consistency provided by the optical flow could be effectively supplemented by modeling in a structural form. This paper proposes a new hierarchical graph neural network (GNN) architecture, dubbed hierarchical graph pattern understanding (HGPU), for zero-shot video object segmentation (ZS-VOS). Inspired by the strong ability of GNNs in capturing structural relations, HGPU innovatively leverages motion cues (i.e., optical flow) to enhance the high-order representations from the neighbors of target frames. Specifically, a hierarchical graph pattern encoder with message aggregation is introduced to acquire different levels of motion and appearance features in a sequential manner. Furthermore, a decoder is designed for hierarchically parsing and understanding the transformed multi-modal contexts to achieve more accurate and robust results. HGPU achieves state-of-the-art performance on four publicly available benchmarks (DAVIS-16, YouTube-Objects, Long-Videos and DAVIS-17). Code and pre-trained model can be found at https://***/NUST-Machine-Intelligence-Laboratory/HGPU.

关键词： Video object segmentation graph neural network zero-shot encoder-decoder optical flow

来源：评论

学校读者我要写书评

暂无评论

Dense Multi-Scale Convolutional Network for Plant Segmentation

引用

IEEE ACCESS 2023年 11卷 82640-82651页

作者： Tran, Thi Hoang Yen Phan, Tran Dang Khoa Univ Danang Fac Econ Univ Econ Da Nang 550000 Vietnam Univ Danang Univ Sci & Technol Fac Elect & Telecommun Engn Da Nang 550000 Vietnam

Plant segmentation is a critical task in precision agriculture as related to crop management and weed treatment. Plants can exhibit very large scale changes, which presents great challenge for accurate crop/weed segmentation. Recent works have shown that multi-scale features are useful to segment objects with different scales. In this work, we propose a Dense Multi-scale Convolutional Network (DMSCN) for pixel-wise crop/weed segmentation. Our network has an encoder-decoder structure. The encoder comprises of a Dense Convolutional Network (DCN) and a Dense Multi-Scale Atrous Pooling (DMSAP) module. DCN is composed of standard and atrous convolutions with dense connections. The architecture of DCN allows the encoder to increase the density of feature maps while avoiding signal decimation due to the dimension reduction. The proposed DMSAP connects a set of standard and atrous convolutional layers with different dilation rates in a densely cascaded manner. DMSAP is able to capture features with dense scale sampling and large receptive field. A simple yet effective decoder is used to refine the segmentation results by combining high and low-level features of the encoder. Extensive experiments are performed on four crop/weed datasets. One of these datasets was collected and annotated by us. We conduct an ablation study to show the advantages of different modules of DMSCN. The comparative study demonstrates the advantages of our model compared with the previous methods in terms of accuracy and complexity.

关键词： INDEX TERMS Atrous convolution dense connection encoder-decoder semantic segmentation spatial pyramid pooling

来源：评论

学校读者我要写书评

暂无评论

Cerebro: Static Subsuming Mutant Selection

引用

IEEE TRANSACTIONS ON SOFTWARE ENGINEERING 2023年第1期49卷 24-43页

作者： Garg, Aayush Ojdanic, Milos Degiovanni, Renzo Chekam, Thierry Titcheu Papadakis, Mike Le Traon, Yves Univ Luxembourg Esch Sur Alzette Luxembourg SES Betzdorf Luxembourg

Mutation testing research has indicated that a major part of its application cost is due to the large number of low utility mutants that it introduces. Although previous research has identified this issue, no previous study has proposed any effective solution to the problem. Thus, it remains unclear how to mutate and test a given piece of code in a best effort way, i.e., achieving a good trade-off between invested effort and test effectiveness. To achieve this, we propose Cerebro, a machine learning approach that statically selects subsuming mutants, i.e., the set of mutants that resides on the top of the subsumption hierarchy, based on the mutants ' surrounding code context. We evaluate Cerebro using 48 and 10 programs written in C and Java, respectively, and demonstrate that it preserves the mutation testing benefits while limiting application cost, i.e., reduces all cost application factors such as equivalent mutants, mutant executions, and the mutants requiring analysis. We demonstrate that Cerebro has strong inter-project prediction ability, which is significantly higher than two baseline methods, i.e., supervised learning on features proposed by state-of-the-art, and random mutant selection. More importantly, our results show that Cerebro 's selected mutants lead to strong tests that are respectively capable of killing 2 times higher than the number of subsuming mutants killed by the baselines when selecting the same number of mutants. At the same time, Cerebro reduces the cost-related factors, as it selects, on average, 68% fewer equivalent mutants, while requiring 90% fewer test executions than the baselines.

关键词： Testing Java Codes Machine learning Costs Supervised learning Reliability Mutant mutation mutation testing subsuming mutant mutant prediction static selection static mutant selection static subsuming mutant selection static subsuming mutant prediction encoder-decoder machine translation tf-seq2seq

来源：评论

学校读者我要写书评

暂无评论

STP-TrellisNets plus : Spatial-Temporal Parallel TrellisNets for Multi-Step Metro Station Passenger Flow Prediction

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2023年第7期35卷 7526-7540页

作者： Ou, Junjie Sun, Jiahui Zhu, Yichen Jin, Haiming Liu, Yijuan Zhang, Fan Huang, Jianqiang Wang, Xinbing Shanghai Jiao Tong Univ Informat & Commun Engn Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ Comp Sci & Technol Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ John Hopcroft Ctr Comp Sci Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ Dept Elect Engn Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ Elect Engn Shanghai 200240 Peoples R China Chinese Acad Sci SIAT Beijing 100045 Peoples R China Alibaba Damo Acad Hangzhou Zhejiang Peoples R China

The drastic increase of metro passengers in recent years inevitably causes the overcrowdedness in the metro systems. Accurately predicting passenger flows at metro stations is critical for efficient metro system management, which helps alleviate such overcrowdedness. Compared to the prevalent next-step prediction, multi-step passenger flow prediction could prominently increase the prediction duration and reveal finer-grained passenger flow variations, which better helps metro system management. Thus, in this paper, we address the problem of multi-step metro station passenger (MSP) flow prediction. In light of MSP flows' unique spatial-temporal characteristics, we propose STP-TrellisNets+, which for the first time augments the newly-emerged temporal convolutional framework TrellisNet for multi-step MSP flow prediction. The temporal module of STP-TrellisNets+ (named CP-TrellisNetsED) employs a Closeness TrellisNet followed by a Periodicity TrellisNets-based encoder-decoder (P-TrellisNetsED) to jointly capture the short- and long-term temporal correlation of MSP flows. In parallel to CP-TrellisNetsED, its spatial module (named GC-TrellisNetsED) adopts a novel transfer flow-based metric to characterize the spatial correlation among MSP flows, and implements another TrellisNetsED on multiple diffusion graph convolutional networks (DGCNs) in time-series order to capture the dynamics of such spatial correlation. Extensive experiments with two large-scale real-world automated fare collection datasets demonstrate that STP-TrellisNets+ outperforms the state-of-the-art baselines.

关键词： Metro station passenger flow multi-step prediction TrellisNet encoder-decoder diffusion graph convolution

来源：评论

学校读者我要写书评

暂无评论

Audio signal Enhancement based on amplitude and phase deep learning model 7

Audio signal Enhancement based on amplitude and phase deep l...

引用

IEEE 7th International Conference on Advanced Technologies, Signal and Image Processing (ATSIP)

作者： Benneila, Mahrez Hadji, Salah ben Messouad, Mouhamed Anouar Natl Engineers Sch Tunis Lab Signal Images & Technol Informat LR SITI ENIT Tunis Tunisia CERTn Borj Hamm Lif Lab Nanomat & Energy Lanser Tunis Tunisia

ISBN: (纸本)9798350351491;9798350351484

In this work, we apply an approach based on simultaneous enhancement of the amplitude and phase time-frequency representation of a composite audio signal with two datasets, namely the Voice Bank+Demand and the NTCD TIMIT datasets. The approach is based on an encoder and a decoder connected by transformers having a convolution module. The encoder uses both amplitude and phase spectra as input, while the decoder has two parts one of parallel amplitude mask and the other is a phase decoder. Thus allowing direct recovery of clean amplitude spectra and also clean enveloped phase spectra, its activation function is a sigmoid learnable and it estimates the phase in parallel. Like all neural networks there are losses at several levels such as: amplitude, phase, and short-term complex, without forgetting the temporal waveforms. All these losses are used jointly in training the model implemented, the validation of this model is done on the two datasets and the experimental results gave a PESQ of 3.5 for the Voice Bank+demand and a PESQ of 3.22 for NTCD-TIMIT.

关键词： speech enhancement encoder-decoder parallel denoising magnitude spectra phase spectra

来源：评论

学校读者我要写书评

暂无评论

A Deep Reinforcement Learning-Based Decision-Making Approach for Routing Problems

引用

APPLIED SCIENCES-BASEL 2025年第9期15卷 4951-4951页

作者： Yan, Dapeng Guan, Qingshu Ou, Bei Yan, Bowen Zhu, Zheng Cao, Hui Xi An Jiao Tong Univ Sch Elect Engn Xian 710049 Peoples R China Xi An Jiao Tong Univ State Key Lab Elect Insulat & Power Equipment Xian 710049 Peoples R China

In recent years, routing problems have attracted significant attention in the fields of operations research and computer science due to their fundamental importance in logistics and transportation. However, most existing learning-based methods employ simplistic context embeddings to represent the routing environment, which constrains their capacity to capture real-time visitation dynamics. To address this limitation, we propose a deep reinforcement learning-based decision-making framework (DRL-DM) built upon an encoder-decoder architecture. The encoder incorporates a batch normalization fronting mechanism and a gate-like threshold block to enhance the quality of node embeddings and improve convergence speed. The decoder constructs a dynamic-aware context embedding that integrates relational information among visited and unvisited nodes, along with the start and terminal locations, thereby enabling effective tracking of real-time state transitions and graph structure variations. Furthermore, the proposed approach exploits the intrinsic symmetry and circularity of routing solutions and adopts an actor-critic training paradigm with multiple parallel trajectories to improve exploration of the solution space. Comprehensive experiments conducted on both synthetic and real-world datasets demonstrate that DRL-DM consistently outperforms heuristic and learning-based baselines, achieving up to an 8.75% reduction in tour length. Moreover, the proposed method exhibits strong generalization capabilities, effectively scaling to larger problem instances and diverse node distributions, thereby highlighting its potential for solving complex, real-life routing tasks.

关键词： attention mechanism deep reinforcement learning encoder-decoder routing

来源：评论

学校读者我要写书评

暂无评论

Retinal vessel segmentation via a Multi-resolution Contextual Network and adversarial learning

引用

NEURAL NETWORKS 2023年第1期165卷 310-320页

作者： Khan, Tariq M. Naqvi, Syed S. Robles-Kelly, Antonio Razzak, Imran Univ New South Wales Sch Comp Sci & Engn Sydney NSW Australia COMSATS Univ Islamabad Dept Elect & Comp Engn Islamabad Pakistan Deakin Univ Fac Sci Engn & Built Environm Sch Informat Technol Locked Bag 20000 Geelong Australia Def Sci & Technol Grp Edinburgh SA 5111 Australia

Timely and affordable computer-aided diagnosis of retinal diseases is pivotal in precluding blindness. Accurate retinal vessel segmentation plays an important role in disease progression and diagnosis of such vision-threatening diseases. To this end, we propose a Multi-resolution Contextual Network (MRC-Net) that addresses these issues by extracting multi-scale features to learn contextual depen-dencies between semantically different features and using bi-directional recurrent learning to model former-latter and latter-former dependencies. Another key idea is training in adversarial settings for foreground segmentation improvement through optimization of the region-based scores. This novel strategy boosts the performance of the segmentation network in terms of the Dice score (and correspondingly Jaccard index) while keeping the number of trainable parameters comparatively low. We have evaluated our method on three benchmark datasets, including DRIVE, STARE, and CHASE, demonstrating its superior performance as compared with competitive approaches elsewhere in the literature.(C) 2023 Elsevier Ltd. All rights reserved.

关键词： Retinal vessel segmentation encoder-decoder Contextual network Adversarial learning Diabetic retinopathy

来源：评论

学校读者我要写书评

暂无评论

Ship Trajectory Prediction for Environmental Perception in Multi-ship Interactions Based on Denoising Diffusion Models 30th

Ship Trajectory Prediction for Environmental Perception in M...

引用

30th International Conference on Computational and Experimental Engineering and Sciences

作者： Ma, Kun Han, Qilong Yao, Jingzheng Harbin Engn Univ Coll Comp Sci & Technol Harbin Peoples R China Harbin Engn Univ Yantai Res Inst Yantai Peoples R China

ISBN: (纸本)9783031816758;9783031816734;9783031816727

Ship trajectory prediction plays a crucial role in maritime safety and navigation, especially in complex environments where multiple ships interact simultaneously. In maritime applications, exact single trajectory prediction may not be paramount;rather, balancing accuracy and diversity of trajectories is essential to ensure collision avoidance based on predicted paths during ship interactions. However, due to the stochastic nature of ships' future intentions, resulting in diverse motion patterns and multiple possible paths, predicting trajectories in these regions poses significant challenges. This paper proposes a novel method, Environmental Perception based on Denoising Diffusion Models(EP-DDM), for ship trajectory prediction in multi-ship interaction scenarios, utilizing denoising diffusion models to enhance environmental perception. EP-DDM is based on an encoder-decoder architecture but differs from conventional methods by avoiding the use of latent variables to represent multimodality. The encoder leverages a transformer to extract historical and interactional information, while the decoder, also based on a transformer, adequately captures temporal dependencies. By integrating denoising techniques into the prediction model, this method aims to improve trajectory prediction accuracy while considering trajectory diversity, particularly in scenes involving complex ship interactions. Experimental results on extensive natural ship trajectory data demonstrate that the proposed model achieves excellent prediction performance and robustness compared to baseline methods, highlighting the potential of our approach to enhance maritime trajectory prediction capabilities in dynamic and congested ocean environments.

关键词： Diffusion model encoder-decoder Trajectory prediction Transfomer

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：