Filters








23,315 Hits in 3.5 sec

Lightweight Prediction and Boundary Attention-based Semantic Segmentation for Road Scene Understanding

Jee-Young Sun, Seung-Won Jung, Sung-Jea Ko
2020 IEEE Access  
Semantic segmentation is one of the most commonly used techniques for road scene understanding.  ...  Recently developed deep learning-based semantic segmentation networks are typically based on the encoder-decoder structure and have made great progress in road scene understanding.  ...  UPPER-LEVEL BOUNDARY ATTENTION In our previous work [55] , BA unit was introduced for binary road segmentation.  ... 
doi:10.1109/access.2020.3001679 fatcat:xb7crjtzlna3jnblsynw22zvuq

The geometric attention-aware network for lane detection in complex road scenes

JianWu Long, ZeRan Yan, Lang Peng, Tong Li, Feng Chen
2021 PLoS ONE  
To solve the problem of lane detection in the various complex road scenes, we proposed a geometric attention-aware network (GAAN) for lane detection.  ...  The proposed GAAN adopted a multi-task branch architecture, and used the attention information propagation (AIP) module to perform communication between branches, then the geometric attention-aware (GAA  ...  Hence the kind of hard classification method is not beneficial for the segmentation of the lane boundary in complex scenes.  ... 
doi:10.1371/journal.pone.0254521 pmid:34264996 pmcid:PMC8282020 fatcat:4zeffsfqrjarxhngpipzlloag4

Methodology for Interactive Labeling of Patched Asphalt Pavement Images Based on U-Net Convolutional Neural Network

Han-Cheng Dan, Hao-Fan Zeng, Zhi-Heng Zhu, Ge-Wen Bai, Wei Cao
2022 Sustainability  
learning combined with reverse and correction labeling.  ...  In this study, in order for an accurate and efficient recognition of the pavement patching images, an interactive labeling method is proposed based on the U-Net convolutional neural network, using active  ...  for threshold segmentation.  ... 
doi:10.3390/su14020861 fatcat:dpelxrjtpjbefl7c7uj3nhjyf4

Uncertainty, Edge, and Reverse-Attention Guided Generative Adversarial Network for Automatic Building Detection in Remotely Sensed Images [article]

Somrita Chattopadhyay, Avinash C. Kak
2021 arXiv   pre-print
The edge attention enhances the boundary features to estimate building boundaries with greater precision, and the reverse attention allows the network to explore the features missing in the previously  ...  To overcome these problems, we propose a generative adversarial network based segmentation framework with uncertainty attention unit and refinement module embedded in the generator.  ...  The Edge Attention Unit amplifies the boundary features, and, thus, helps the network to learn precise boundaries of the buildings.  ... 
arXiv:2112.05335v1 fatcat:45gluyizlneqvbp3dyeh7jg4vq

Uncertainty, Edge, and Reverse-Attention Guided Generative Adversarial Network for Automatic Building Detection in Remotely Sensed Images

Somrita Chattopadhyay, Avinash C. Kak
2022 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
The edge attention enhances the boundary features to estimate building boundaries with greater precision, and the reverse attention allows the network to explore the features missing in the previously  ...  To overcome these problems, we propose a generative adversarial network-based segmentation framework with uncertainty attention unit and refinement module embedded in the generator.  ...  Another novel aspect of our network is the RM that consists of a reverse-attention unit (RAU) and an edge attention unit (EAU).  ... 
doi:10.1109/jstars.2022.3166929 fatcat:ugcdx73ujbhxfdxhkghczbkvue

RoadNet-RT: High Throughput CNN Architecture and SoC Design for Real-Time Road Segmentation [article]

Lin Bai, Yecheng Lyu, Xinming Huang
2021 arXiv   pre-print
It achieves 90.33% MaxF score on test set of KITTI road segmentation task and 8 ms per frame when running on GTX 1080 GPU.  ...  In order to reach real-time process speed, a light-weight, high-throughput CNN architecture namely RoadNet-RT is proposed for road segmentation in this paper.  ...  Instead of using the direct skip connection in U-Net [7] and SegNet [8] , a residual refinement module bridged encoder and decoder parts, which consisted of reversed attention and boundary attention  ... 
arXiv:2006.07644v2 fatcat:gwo5vub4rbbc3fzqnz6xnswj2e

A pruning algorithm for reverse nearest neighbors in directed road networks

Rizwan Qamar, Muhammad Attique, Tae-Sun Chung
2015 2015 IEEE/ACIS 14th International Conference on Computer and Information Science (ICIS)  
Although, much research has been done for RNN in Euclidean and undirected network space, very less attention has been paid to directed road network, where network distances are not symmetric.  ...  In this paper, we studied the problem of continuous reverse k nearest neighbors (RkNN) in directed road network, where a road segment can have a particular orientation.  ...  Safe exit points are the boundary points of the safe region and since safe region is comprised of road segments and safe exit points are just points in the road network, as a result less network bandwidth  ... 
doi:10.1109/icis.2015.7166606 dblp:conf/ACISicis/QamarAC15 fatcat:xkwakwuzpfdk7jy642tcumghxu

CaraNet: Context Axial Reverse Attention Network for Segmentation of Small Medical Objects [article]

Ange Lou, Shuyue Guan, Hanseok Ko, Murray Loew
2022 arXiv   pre-print
Recently, many convolutional neural networks (CNN) have been designed for segmentation tasks and achieved great success.  ...  Segmenting medical images accurately and reliably is important for disease diagnosis and treatment.  ...  Selective feature aggregation network with area- boundary constraints for polyp segmentation.  ... 
arXiv:2108.07368v3 fatcat:w2xqgcckvjdj5cvynu5s3xhbha

Revise-Net: Exploiting Reverse Attention Mechanism for Salient Object Detection

Rukhshanda Hussain, Yash Karbhari, Muhammad Fazal Ijaz, Marcin Woźniak, Pawan Kumar Singh, Ram Sarkar
2021 Remote Sensing  
Finally, multiple reverse attention modules at varying scales are cascaded between the two networks to guide the prediction module by employing the intermediate segmentation maps generated at each downsampling  ...  The proposed Revise-Net model is divided into three parts: (a) the prediction module, (b) a residual enhancement module, and (c) reverse attention modules.  ...  Attentive Feedback Network for Boundary-aware Salient Object Detection.  ... 
doi:10.3390/rs13234941 fatcat:4jno22evrvehbm4zznwfi43yp4

Geometric Constrained Joint Lane Segmentation and Lane Boundary Detection [chapter]

Jie Zhang, Yi Xu, Bingbing Ni, Zhenyu Duan
2018 Lecture Notes in Computer Science  
The existing approaches for lane detection can be categorized as lane area segmentation and lane boundary detection.  ...  In this paper, we establish a multiple-task learning framework to segment lane areas and detect lane boundaries simultaneously.  ...  Given the lane boundaries, segmentation algorithms are applied to the bounded regions to refine lane labels. Reversely, some models segment lane areas first.  ... 
doi:10.1007/978-3-030-01246-5_30 fatcat:dcnbjgj3zngbtl5ftwqcgmkloe

Street Network Studies: from Networks to Models and their Representations

Stephen Marshall, Jorge Gil, Karl Kropf, Martin Tomko, Lucas Figueiredo
2018 Networks and Spatial Economics  
It suggests greater attention to consistent use of terms and concepts, of graph representations and practical applications, and concludes with suggestions for possible ways forward.  ...  The motivations for adopting a given model of the network are also not always clear and often seem to follow disciplinary traditions.  ...  reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.  ... 
doi:10.1007/s11067-018-9427-9 fatcat:afbmi26yfrgvjfuakejxgyhqua

High-Resolution Neural Network for Driver Visual Attention Prediction

Byeongkeun Kang, Yeejin Lee
2020 Sensors  
Second, attention prediction tends to be biased toward centers of images when neural networks are trained using typical visual attention datasets.  ...  Understanding a driver's visual attention and relevant behavior information is a challenging but essential task in advanced driver-assistance systems (ADAS) and efficient autonomous vehicles (AV).  ...  by adapting the convolutional neural networks for semantic segmentation [10] .  ... 
doi:10.3390/s20072030 pmid:32260397 fatcat:vxtepwn2jvg6dg72gvld2wfz7y

Lane and Road Marker Semantic Video Segmentation Using Mask Cropping and Optical Flow Estimation

Guansheng Xing, Ziming Zhu
2021 Sensors  
We segmented lanes and road markers on the Baidu Apolloscape lanemark segmentation dataset and CULane dataset, and present benchmarks for different networks.  ...  single-output network to jointly process the segmentation of lanes and road markers.  ...  [28] introduced an efficient concurrent attention network called efficient concurrent attention networks (ECANets), which include a horizontal segment attention (HSA) module and a pyramidal space attention  ... 
doi:10.3390/s21217156 pmid:34770463 pmcid:PMC8587959 fatcat:brjgel33jnhqhpxvq3gx5qsgpe

Non-Connective Linear Cartograms for Mapping Traffic Conditions

Yi-Hwa Wu, Ming-Chih Hung
2010 Cartographic Perspectives  
All three visual variables work together to create dramatic visual effects and attract greater attention from readers.  ...  During highly congested hours, travel speeds are low, and travel times are long, and vice versa. In previous studies, traffic conditions were visualized by color and width of road segments.  ...  The street network in Salt Lake City, Utah, is a grid-like network where most road segments are short and straight and are well aligned in an east-west or north-south direction.  ... 
doi:10.14714/cp65.128 fatcat:ws5qht5tzjbkznbkd2fmwwf6va

Multi-Stage Improved Route Planning Approach: theoretical foundations [article]

Petr Hlineny, Ondrej Moris
2011 arXiv   pre-print
We provide a theoretical proof that our approach efficiently computes exact optimal routes within this concept, as well as we demonstrate with experimental results on publicly available road networks of  ...  A new approach to the static route planning problem, based on a multi-staging concept and a scope notion, is presented.  ...  The collected data are briefly summarized in Table 4 and Table 3 . To get a taste of the topic, some examples of local cell partitioning results are depicted in Fig. 2 and 1.  ... 
arXiv:1101.3182v2 fatcat:fwnmw6semvbz5kn7rt22uahmhq
« Previous Showing results 1 — 15 out of 23,315 results