检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Lu, Yuwu Lang, Jicong Ding, Meirong School of Artificial Intelligence South China Normal University Foshan China

ISBN: (纸本)9798350368741

nighttime semantic segmentation is an indispensable component in practical applications, such as automated vehicles. However, it is often hindered by the lack of annotations due to interference caused by inadequate lighting or exposure. To overcome these difficulties, we propose a Dual-Path Consistency (DPC) unsupervised domain adaptation (UDA) approach. One path is Image Darkening Path (IDP), in which feature representations of original images and darkened images extracted from the feature encoder are leveraged to maintain cross-domain style consistency. Another path is Image Masking Path (IMP), in which the masked images are reconstructed under the guidance of pseudo-labels, aiming to maintain content consistency in an entirely identical scenario. Extensive experiments on four benchmarks demonstrate the superior performance of the proposed DPC for nighttime semantic segmentation. © 2025 IEEE.

关键词： image masking nighttime semantic segmentation self-training unsupervised domain adaptation

来源：评论

学校读者我要写书评

暂无评论

EGBNet: a real-time edge-guided bilateral network for nighttime semantic segmentation

引用

SIGNAL IMAGE AND VIDEO PROCESSING 2023年第6期17卷 3173-3181页

作者： An, Guanhua Guo, Jichang Wang, Yudong Ai, Yufeng Tianjin Univ Sch Elect & Informat Engn Tianjin Peoples R China

Due to poor illumination and low contrast, semantic segmentation of nighttime images faces major challenges. Various segmentation models with a large number of parameters are proposed to improve the performance but lead to an inability to process in real time. To tackle these problems, we propose a real-time edge-guided bilateral network (EGBNet) for nighttime semantic segmentation. Considering the blurred details and low contrast of nighttime images, we propose a lightweight multi-dilation dense aggregation module and introduce an efficient edge head to improve the ability to distinguish target features from the nighttime background. Moreover, a self-adaptive feature fusion module is proposed for the bilateral segmentation network to enhance the feature representation and generalization ability by fully using multi-scale feature maps. To capture more useful information from limited nighttime images, we further use the knowledge distillation strategy to improve the segmentation performance. Extensive experiments on ACDC and BDD datasets demonstrate the effectiveness of our EGBNet by achieving a satisfactory trade-off between segmentation accuracy and inference speed. Specifically, EGBNet achieves 55.56% mIoU on the ACDC test set with 9.4 M parameters and 60FPS speed for a 1080 x 1920 input image on a single NVIDIA 2080Ti.

关键词： nighttime semantic segmentation Real-time processing Edge-guided Deep learning

来源：评论

学校读者我要写书评

暂无评论

Learning nighttime semantic segmentation the Hard Way

引用

ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS 2024年第7期20卷 1-23页

作者： Liu, Wenxi Cai, Jiaxin Li, Qi Liao, Chenyang Cao, Jingjing He, Shengfeng Yu, Yuanlong Fuzhou Univ Coll Comp & Data Sci 2 Xue Yuan Rd Fuzhou 350108 Fujian Peoples R China Wuhan Univ Technol Sch Transportat & Logist Engn 122 Luoshi Rd Wuhan Hubei Peoples R China Singapore Management Univ Sch Comp & Informat Syst Adm Bldg81 Victoria St Singapore 188065 Singapore

nighttime semantic segmentation is an important but challenging research problem for autonomous driving. The major challenges lie in the small objects or regions from the under-/over-exposed areas or suffer from motion blur caused by the camera deployed on moving vehicles. To resolve this, we propose a novel hard-classaware module that bridges the main network for full-class segmentation and the hard-class network for segmenting aforementioned hard-class objects. In specific, it exploits the shared focus of hard-class objects from the dual-stream network, enabling the contextual information flow to guide the model to concentrate on the pixels that are hard to classify. In the end, the estimated hard-class segmentation results will be utilized to infer the final results via an adaptive probabilistic fusion refinement scheme. Moreover, to overcome over-smoothing and noise caused by extreme exposures, our model is modulated by a carefully crafted pretext task of constructing an exposure-aware semantic gradient map, which guides the model to faithfully perceive the structural and semantic information of hard-class objects while mitigating the negative impact of noises and uneven exposures. In experiments, we demonstrate that our unique network design leads to superior segmentation performance over existing methods, featuring the strong ability of perceiving hard-class objects under adverse conditions.

关键词： nighttime semantic segmentation dual-branch hard class pretext task fusion refinement scheme

来源：评论

学校读者我要写书评

暂无评论

SGDA: A Saliency-Guided Domain Adaptation Network for nighttime semantic segmentation 6

SGDA: A Saliency-Guided Domain Adaptation Network for Nightt...

引用

IEEE 6th International Conference on Industrial Cyber-Physical Systems (ICPS)

作者： Duan, Yijia Tu, Jingzheng Chen, Cailian Shanghai Jiao Tong Univ Dept Automat Key Lab Syst Control & Informat Proc Minist Educ China Shanghai 200240 Peoples R China

ISBN: (纸本)9798350311259

nighttime semantic segmentation has attracted considerable attention due to its crucial status in the smart city. However, it is challenging to handle poor illumination and indiscernible information. To tackle these problems, a saliencyguided domain adaptation network, SGDA, is proposed via adapting daytime models to nighttime scenes. Firstly, a saliency guidance branch is attached to the segmentation network to enrich the spatial features and guide the model to better perceive detail information. Secondly, to embed the saliency guidance to the segmentation network, a pyramid attention architecture is designed to fuse the features from the two branches. Thirdly, an illumination adaptation module is constructed to close the intensity distributions via adversarial learning, with an elaborately designed loss function to improve the performance. Extensive experiments on Dark Zurich dataset and nighttime Driving dataset validate the effectiveness of SGDA, and indicate that our method improves the accuracy on small object categories.

关键词： nighttime semantic segmentation deep learning smart city

来源：评论

学校读者我要写书评

暂无评论

A Two-Stage Self-Training Framework for nighttime semantic segmentation 38

A Two-Stage Self-Training Framework for Nighttime Semantic S...

引用

38th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2023

作者： Yang, Dan Liu, Cuiwei Jiang, Youzhi Shenyang Aerospace University School of Computer Science Shenyang China

ISBN: (纸本)9798350303636

nighttime semantic segmentation is a challenging task due to the poor lighting in the night scenes. Since it is difficult to acquire pixel-level annotations of nighttime images, the existing works attempt to adapt a model learned on annotated daytime images to unlabeled nighttime images. Due to the gap between daytime and nighttime images, the model tends to over-fit daytime images and learn wrong annotation patterns for overexposed regions in nighttime images. To handle these issues, we propose a two-stage self-training framework, which synthesizes high-quality pseudo-labels for nighttime images based on overexposure detection. The synthetic pseudo-labels are then utilized as supervision information to retrain a new segmentation model, which significantly mitigates labeling errors in overexposed regions. Experiments on the DarkZurich dataset verify the effectiveness of the proposed method. © 2023 IEEE.

关键词： nighttime semantic segmentation pseudo-labels self-training

来源：评论

学校读者我要写书评

暂无评论

Frequency-based dual attention network for nighttime semantic segmentation

引用

JOURNAL OF ELECTRONIC IMAGING 2024年第2期33卷 023042-023042页

作者： Wan, Haifeng Guo, Jichang An, Guanhua Zheng, Sida Wang, Yudong Tianjin Univ Sch Elect Automat & Informat Engn Tianjin Peoples R China Shanghai Artificial Intelligence Lab Shanghai Peoples R China

nighttime images suffer from exposure imbalance, leading to significant challenges in nighttime semantic segmentation. Given this, we propose a frequency-based dual attention network (FDANet) for nighttime semantic segmentation. To address the problem of targets hidden in the background caused by uneven exposure and low contrast that are difficult to distinguish, we propose a multi-spectral dual attention module to deeply explore hidden features from frequency information. In addition, to further utilize the relationship between frequency domain information and spatial domain features and guide the extraction of contextual features, spatial-frequency aggregation module is proposed to align and fuse features from different domains. Moreover, self-attention dilated convolution pyramid module is proposed to expand the effective receptive field and enhance the ability to identify multi-scale targets under uneven exposure. Extensive experiments on NightCity, ACDC, and BDD show that FDANet achieves an impressive balance between accuracy and speed. Specifically, FDANet achieves 53.76% MIoU, 49.47% MIoU, and 39.70% MIoU on NightCity, ACDC-Night, and BDD-Night, respectively. Notably, it attains these results with remarkable processing speeds of 111 FPS, 66 FPS, and 104 FPS for input resolutions of 512x1024, 1080x1920, and 720x1280, respectively.

关键词： nighttime semantic segmentation frequency domain multi-scale feature fusion self-attention

来源：评论

学校读者我要写书评

暂无评论

NoctuDroneNet: Real-Time semantic segmentation of nighttime UAV Imagery in Complex Environments

引用

DRONES 2025年第2期9卷 97-97页

作者： Qu, Ruokun Tan, Jintao Liu, Yelu Li, Chenglong Jiang, Hui Civil Aviat Flight Univ China Coll Air Traff Management Guanghan 618307 Peoples R China Univ Elect Sci & Technol China Sch Comp Sci & Engn Chengdu 611731 Peoples R China

nighttime semantic segmentation represents a challenging frontier in computer vision, made particularly difficult by severe low-light conditions, pronounced noise, and complex illumination patterns. These challenges intensify when dealing with Unmanned Aerial Vehicle (UAV) imagery, where varying camera angles and altitudes compound the difficulty. In this paper, we introduce NoctuDroneNet (Nocturnal UAV Drone Network, hereinafter referred to as NoctuDroneNet), a real-time segmentation model tailored specifically for nighttime UAV scenarios. Our approach integrates convolution-based global reasoning with training-only semantic alignment modules to effectively handle diverse and extreme nighttime conditions. We construct a new dataset, NUI-Night, focusing on low-illumination UAV scenes to rigorously evaluate performance under conditions rarely represented in standard benchmarks. Beyond NUI-Night, we assess NoctuDroneNet on the Varied Drone Dataset (VDD), a normal-illumination UAV dataset, demonstrating the model's robustness and adaptability to varying flight domains despite the lack of large-scale low-light UAV benchmarks. Furthermore, evaluations on the Night-City dataset confirm its scalability and applicability to complex nighttime urban environments. NoctuDroneNet achieves state-of-the-art performance on NUI-Night, surpassing strong real-time baselines in both segmentation accuracy and speed. Qualitative analyses highlight its resilience to under-/over-exposure and small-object detection, underscoring its potential for real-world applications like UAV emergency landings under minimal illumination.

关键词： nighttime semantic segmentation UAV imagery NoctuDroneNet low-illumination conditions real-time segmentation NUI-Night dataset

来源：评论

学校读者我要写书评

暂无评论

semantic nighttime IMAGE segmentation VIA ILLUMINATION AND POSITION AWARE DOMAIN ADAPTATION

SEMANTIC NIGHTTIME IMAGE SEGMENTATION VIA ILLUMINATION AND P...

引用

IEEE International Conference on Image Processing (ICIP)

作者： Peng, Junhan Su, Jia Sun, Yongqing Wang, Zheng Lin, Chia-wen Capital Normal Univ Beijing Peoples R China Nippon Telegraph & Tel Corp Yokosuka Kanagawa Japan Wuhan Univ Wuhan Peoples R China Natl Tsing Hua Univ Hsinchu Taiwan

ISBN: (纸本)9781665441155

Due to the lack of the annotated nighttime images, general image segmentation models trained on the daytime image dataset do not perform well in nighttime scenes. The difference of the illumination condition and the difficulty to obtain the position information between daytime and nighttime makes the nighttime image segmentation tough. As a consequence, this paper proposes an end-to-end nighttime segmentation network based on the following two points: 1) Utilizing illumination adaptation with the different illumination condition on the daytime or nighttime to close the distribution gap at the feature map level;2) With the prior information about the position of each object in the outdoor scene, some classification errors could be corrected by incorporating the self-attention mechanism. The scheme is tested on the open-source nighttime dataset Dark Zurich and night driving, with a 2.5% improvement compared to the base segmentation network.

关键词： nighttime semantic segmentation Domain Adaptation Self-attention

来源：评论

学校读者我要写书评

暂无评论

Meta-Learning Based Knowledge Distillation for Domain Adaptive nighttime segmentation 7th

Meta-Learning Based Knowledge Distillation for Domain Adapti...

引用

7th Chinese Conference on Pattern Recognition and Computer Vision

作者： Guan, Hao Liu, Jun Wang, Simiao Li, Yunan Lu, Mingyu Univ Sci & Technol China Sch Software Engn Hefei 230000 Peoples R China Dalian Maritime Univ Sch Artificial Intelligence Dalian 116024 Peoples R China Qilu Univ Technol Key Lab Comp Power Network & Informat Secur Minist Educ Shandong Acad Sci Jinan 250353 Peoples R China

ISBN: (纸本)9789819784899;9789819784905

semantic segmentation of nighttime scenes poses a significant challenge in autonomous driving. While unsupervised domain adaptation offers an effective solution, existing methods disregard the relationship of knowledge transfer across different domains, which is crucial for improving model generalization ability. In this paper, we propose a Meta-Learning based Knowledge Distillation (MLKD) method tailored for adapting models trained on a source domain (daytime scene) to target domains (nighttime scenes). Firstly, we propose a brightness adjustment module based on the fast Fourier transform, which generates source images resembling the target scene in the latent domain without additional training burden. Secondly, we introduce a mask-based consistency constraint to extract knowledge from complementary latent images. This enables the model to capture rich spatial contextual relationships in scenes with lighting variations, resulting in more compact representations. Thirdly, we construct a bi-level meta-learning framework that transfers cross-domain knowledge learned from the pair of "source-to-latent" to enhance the adaptation of "latent-to-target". Extensive experiments on benchmark datasets, i.e., Dark Zurich and ACDC, show that our MLKD achieves state-of-the-art performance, demonstrating the effectiveness of our approach in nighttime semantic segmentation.

关键词： nighttime semantic segmentation Unsupervised domain adaptation Brightness adjustment module Meta-learning

来源：评论

学校读者我要写书评

暂无评论

Complementary Masked-Guided Meta-Learning for Domain Adaptive nighttime segmentation

引用

IEEE SIGNAL PROCESSING LETTERS 2024年 31卷 3010-3014页

作者： Chen, Ruiying Bo, Yuming Wu, Panlong Wang, Simiao Liu, Yunan Nanjing Univ Sci & Technol Sch Automat Nanjing 210094 Peoples R China Dalian Maritime Univ Sch Artificial Intelligence Dalian 116026 Peoples R China

semantic segmentation in nighttime scenes presents a significant challenge in autonomous driving. Unsupervised domain adaptation (UDA) offers an effective solution by learning domain-invariant features to transfer models from the source domain (daytime scenes) to the target domain (nighttime scenes). Many methods introduce a latent domain to reduce the difficulty of UDA. However, they often build only a single adaptation pair of "latent-to-target", which limits the effectiveness of knowledge transfer across different domains. In this letter, we propose a Masked Guided Meta-Learning (MGML) framework for domain-adaptive nighttime semantic segmentation. Within the MGML framework, we explore two key issues: how to generate the latent domain, and how to leverage the latent domain to assist meta-learning in reducing domain discrepancy. For the first issue, we employ the fast Fourier transform along with a complementary masking strategy to generate masked latent images that resemble the target scenes in the latent domain without adding to the training burden. For the second issue, we nest a mask-based consistency constraint within a bi-level meta-learning framework, enabling cross-domain knowledge acquired from the pair of "source-to-latent" to enhance the "latent-to-target" adaptation. Experiments on benchmark datasets demonstrate that our MGML achieves state-of-the-art performance, demonstrating the effectiveness of our approach in nighttime semantic segmentation.

关键词： Metalearning Adaptation models semantic segmentation Training Optimization Fast Fourier transforms semantics nighttime semantic segmentation mask-based consistency constraint meta-learning

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：