Filters








8,257 Hits in 4.2 sec

Vision for Autonomous Vehicles and Probes (Dagstuhl Seminar 15461)

André Bruhn, Atsushi Imiya, Ales Leonardis, Tomas Pajdla, Marc Herbstritt
2016 Dagstuhl Reports  
Continuing topics of interest in computer vision are scene and environmental understanding using singleand multiple-camera systems, which are fundamental techniques for autonomous driving, navigation in  ...  as the central component for autonomous driving and navigation and remote exploration.  ...  for Autonomous Driving Andreas Geiger High-level Knowledge in Low-level Vision Bernt Schiele Towards 3D Scene Understanding Vision for Motion Analysis Cédric Demonceaux Pose Estimation and 3D Segmentation  ... 
doi:10.4230/dagrep.5.11.36 dblp:journals/dagstuhl-reports/BruhnILP15 fatcat:l2nqd45tnrabpdqmwex6enkxei

Vision-Based Offline-Online Perception Paradigm for Autonomous Driving

German Ros, Sebastian Ramos, Manuel Granados, Amir Bakhtiary, David Vazquez, Antonio M. Lopez
2015 2015 IEEE Winter Conference on Applications of Computer Vision  
Autonomous driving is a key factor for future mobility.  ...  Then, detecting the dynamic obstacles we obtain a rich understanding of the current scene.  ...  In order to address the challenge of understanding urban scenes, most research has focused in scene semantic segmentation and 3D mapping as partial solutions.  ... 
doi:10.1109/wacv.2015.38 dblp:conf/wacv/RosRGBVL15 fatcat:m2hfogdmmffmhpmhibqahd2ouy

Ground-distance segmentation of 3D LiDAR point cloud toward autonomous driving

Jian Wu, Qingxiong Yang
2020 APSIPA Transactions on Signal and Information Processing  
In this paper, we study the semantic segmentation of 3D LiDAR point cloud data in urban environments for autonomous driving, and a method utilizing the surface information of the ground plane was proposed  ...  This paper is focusing on semantic segmentation of the sparse point clouds obtained from 32-channel LiDAR sensor with deep neural networks.  ...  This section briefly discusses some recent works in dynamic outdoor scenes as follows: • Scene understanding in autonomous drivingSemantic segmentation of point clouds • Semantic segmentation of large-scale  ... 
doi:10.1017/atsip.2020.21 fatcat:ihfiaqckmfaghoxreojsgvoorm

Multi-modal Sensor Fusion-Based Deep Neural Network for End-to-end Autonomous Driving with Scene Understanding [article]

Zhiyu Huang, Chen Lv, Yang Xing, Jingda Wu
2020 arXiv   pre-print
The designed end-to-end deep neural network takes the visual image and associated depth information as inputs in an early fusion level and outputs the pixel-wise semantic segmentation as scene understanding  ...  This study aims to improve the control performance and generalization capability of end-to-end autonomous driving with scene understanding leveraging deep learning and multimodal sensor fusion technology  ...  Here we choose α = 5, β = 1, and γ = 2 for Eq. 6. For scene understanding learning, we use cross-entropy loss (Eq. 8) for the semantic segmentation task. D.  ... 
arXiv:2005.09202v2 fatcat:fnr5xcrsvvflfikly7tejhvgpu

S3-Net: A Fast and Lightweight Video Scene Understanding Network by Single-shot Segmentation [article]

Yuan Cheng, Yuchao Yang, Hai-Bao Chen, Ngai Wong, Hao Yu
2020 arXiv   pre-print
Real-time understanding in video is crucial in various AI applications such as autonomous driving. This work presents a fast single-shot segmentation strategy for video scene understanding.  ...  The proposed net, called S3-Net, quickly locates and segments target sub-scenes, meanwhile extracts structured time-series semantic features as inputs to an LSTM-based spatio-temporal model.  ...  it a strong candidate for real-time video scene understanding in autonomous driving.  ... 
arXiv:2011.02265v1 fatcat:xifjlxrxu5atbb5zjkas7jyoeu

A Survey on Deep Learning Based Approaches for Scene Understanding in Autonomous Driving

Zhiyang Guo, Yingping Huang, Xing Hu, Hongjian Wei, Baigan Zhao
2021 Electronics  
As a prerequisite for autonomous driving, scene understanding has attracted extensive research.  ...  This paper aims to provide a comprehensive survey of deep learning-based approaches for scene understanding in autonomous driving.  ...  in Autonomous Driving In this section, we review deep learning-based approaches for scene understanding in terms of four work streams: object detection, full-scene semantic segmentation, instance segmentation  ... 
doi:10.3390/electronics10040471 fatcat:gyloykg24nbqvlw4ujiiagoneq

A Fine-Grained Dataset and its Efficient Semantic Segmentation for Unstructured Driving Scenarios [article]

Kai A. Metzger, Peter Mortimer, Hans-Joachim Wuensche
2021 arXiv   pre-print
In this paper, we introduce TAS500, a novel semantic segmentation dataset for autonomous driving in unstructured environments.  ...  Research in autonomous driving for unstructured environments suffers from a lack of semantically labeled datasets compared to its urban counterpart.  ...  INTRODUCTION Semantic scene understanding is a key capability for autonomous robot navigation in real-world environments, but current research in autonomous driving focuses mainly on urban, suburban, and  ... 
arXiv:2103.13109v1 fatcat:owxxlaehlfa25khlcusqfoqsna

Autonomous driving: cognitive construction and situation understanding

Shitao Chen, Zhiqiang Jian, Yuhao Huang, Yu Chen, Zhuoli Zhou, Nanning Zheng
2019 Science China Information Sciences  
In addition, for the process of understanding traffic scenes, the efficiency of event-related mechanism is more significant than single-attribute scenario perception data.  ...  In current research of autonomous driving, the most widely adopted technique is to use a basic framework of serial information processing and computations, which consists of four modules: perception, planning  ...  The semantic segmentation of traffic scenes is used to distinguish pixel category information of different categories in an image.  ... 
doi:10.1007/s11432-018-9850-9 fatcat:qys3uucz3zgznfou6vgfjerwlq

PSI: A Pedestrian Behavior Dataset for Socially Intelligent Autonomous Car [article]

Tina Chen, Renran Tian, Yaobin Chen, Joshua Domeyer, Heishiro Toyoda, Rini Sherony, Taotao Jing, Zhengming Ding
2021 arXiv   pre-print
Prediction of pedestrian behavior is critical for fully autonomous vehicles to drive in busy city streets safely and efficiently.  ...  These innovative labels can enable several computer vision tasks, including pedestrian intent/behavior prediction, vehicle-pedestrian interaction segmentation, and video-to-language mapping for explainable  ...  Understanding pedestrian behavior in complex traffic scenes.  ... 
arXiv:2112.02604v1 fatcat:znovxwe5gjhnhczjx7ximrs5qu

Using Image Priors to Improve Scene Understanding [article]

Brigit Schroeder, Hanlin Tang, Alexandre Alahi
2019 arXiv   pre-print
Semantic segmentation algorithms that can robustly segment objects across multiple camera viewpoints are crucial for assuring navigation and safety in emerging applications such as autonomous driving.  ...  We propose a simple yet effective method for leveraging these image priors to improve semantic segmentation of images from sequential driving datasets.  ...  This leads us to believe that scene graphs would be effective form of prior knowledge to improve as semantic segmentation and potentially other scene understanding algorithms.  ... 
arXiv:1910.01198v1 fatcat:xxa2qptdhraidfdr5n6e7meepe

The ApolloScape Open Dataset for Autonomous Driving and its Application [article]

Xinyu Huang, Peng Wang, Xinjing Cheng, Dingfu Zhou, Qichuan Geng, Ruigang Yang
2018 arXiv   pre-print
for autonomous driving.  ...  In this paper, we present the ApolloScape dataset [1] and its applications for autonomous driving. Compared with existing public datasets from real scenes, e.g.  ...  We also thank the work of Xibin Song, Binbin Cao, Jin Fang, He Jiang, Yu Zhang, Xiang Gu, and Xiaofei Liu for their laborious efforts in organizing data, helping writing label tools, checking labelled  ... 
arXiv:1803.06184v3 fatcat:l42yzyondnartoldw2orr24hly

A Methodological Review of Visual Road Recognition Procedures for Autonomous Driving Applications [article]

Kai Li Lim, Thomas Bräunl
2019 arXiv   pre-print
road recognition, with an emphasis on methods that incorporate convolutional neural networks and semantic segmentation.  ...  In order for vehicles to be fully autonomous, it is imperative that the driver assistance system is adapt in road and lane keeping.  ...  Acknowledgment The authors would like to thank Mr Andrea Palazzi, Mr Thomas Anthony, Mr Touqeer Ahmad and Prof Hui Kong for the granting of permission to use their figures in this paper.  ... 
arXiv:1905.01635v1 fatcat:l26j7p665va4dkiplnj5gntxoa

Advanced Visual Analyses for Smart and Autonomous Vehicles

Zhijun Fang, Jenq-Neng Hwang, Shih-Chia Huang
2018 Advances in Multimedia  
Adversarial Networks" proposes a scene understanding framework based on a generative adversarial network (GAN) to implement a fully convolutional semantic segmentation model. e highorder potentials are  ...  A er several iterations of reviewing processes, five papers are accepted for this special issue, which covers the advance of visual analysis techniques for visual tracking, scene understanding, lane detection  ... 
doi:10.1155/2018/1762428 fatcat:zfhljk3hdndxbnmej2zzgg56ry

Scene Completeness-Aware Lidar Depth Completion for Driving Scenario [article]

Cho-Ying Wu, Ulrich Neumann
2021 arXiv   pre-print
Moreover, we experiment on less-explored outdoor RGBD semantic segmentation with scene completeness-aware D-input to validate our method.  ...  These areas are considered less important since they are usually sky or trees of less scene understanding interest.  ...  We use SSMA [15] , a high-performing outdoor RGB-D semantic segmentation framework, as the baseline to show that our recovered depth could help better scene understanding.  ... 
arXiv:2003.06945v3 fatcat:sne365a2s5bvtaumx5zrcmeiym

BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning [article]

Fisher Yu, Haofeng Chen, Xin Wang, Wenqi Xian, Yingying Chen, Fangchen Liu, Vashisht Madhavan, Trevor Darrell
2020 arXiv   pre-print
Datasets drive vision progress, yet existing driving datasets are impoverished in terms of visual content and supported tasks to study multitask learning for autonomous driving.  ...  We construct BDD100K, the largest driving video dataset with 100K videos and 10 tasks to evaluate the exciting progress of image recognition algorithms on autonomous driving.  ...  Object Detection Locating objects is a fundamental task for not only autonomous driving but the general visual recognition.  ... 
arXiv:1805.04687v2 fatcat:7q4hsr2435eqzlz4rzienja2vm
« Previous Showing results 1 — 15 out of 8,257 results