Filters








2,041 Hits in 4.3 sec

Detecting Ground Control Points via Convolutional Neural Network for Stereo Matching [article]

Zhun Zhong, Songzhi Su, Donglin Cao, Shaozi Li
2016 arXiv   pre-print
In this paper, we present a novel approach to detect ground control points (GCPs) for stereo matching problem.  ...  First of all, we train a convolutional neural network (CNN) on a large stereo set, and compute the matching confidence of each pixel by using the trained CNN model.  ...  Conclusion This paper presents a Convolutional Neural Network based approach that is able to detect the ground control points (GCPs) according to the matching confidence of each pixel.  ... 
arXiv:1605.02289v1 fatcat:tlpbterejffehfkbb47r5kqeti

Sparse Coding on Stereo Video for Object Detection [article]

Sheng Y. Lundquist, Melanie Mitchell, Garrett T. Kenyon
2017 arXiv   pre-print
Deep Convolutional Neural Networks (DCNN) require millions of labeled training examples for image classification and object detection tasks, which restrict these models to domains where such datasets are  ...  We show that replacing a typical supervised convolutional layer with an unsupervised sparse-coding layer within a DCNN allows for better performance on a car detection task when only a limited number of  ...  Paiton for valuable conversations pertaining to this manuscript.  ... 
arXiv:1705.07144v2 fatcat:mnox5dgizzaqbmzah675yyxlca

Multi-Attention Network for Stereo Matching

Xiaowei Yang, Lin He, Yong Zhao, Haiwei Sang, Zuliu Yang, Xianjing Cheng
2020 IEEE Access  
To make better use the global context information for stereo matching, we propose a novel convolutional neural network.  ...  RELATED WORK There are many studies on stereo matching. We learn from methods employing convolutional neural networks.  ...  XIAOWEI YANG received his undergraduate degree in Measurement and control technology and instrument specialty from Nanyang Institute of Technology, Nanyang, China, in 2013. He  ... 
doi:10.1109/access.2020.3003375 fatcat:k466odrmnjeflhqrnmx4fsmsaq

End-to-end Learning of Cost-Volume Aggregation for Real-time Dense Stereo [article]

Andrey Kuzmin, Dmitry Mikushin, Victor Lempitsky
2016 arXiv   pre-print
We present a new deep learning-based approach for dense stereo matching.  ...  At the same time, our approach uses a deep convolutional network to predict the local parameters of cost volume aggregation process, which in this paper we implement using differentiable domain transform  ...  CONCLUSION We proposed a new method of computing dense stereo correspondences using convolutional neural network trained to aggregate the cost volume.  ... 
arXiv:1611.05689v1 fatcat:4y7sg7e2qvdxvkj4gx6elwyyf4

MSDC-Net: Multi-Scale Dense and Contextual Networks for Automated Disparity Map for Stereo Matching [article]

Zhibo Rao and Mingyi He and Yuchao Dai and Zhidong Zhu and Bo Li and Renjie He
2019 arXiv   pre-print
Disparity prediction from stereo images is essential to computer vision applications including autonomous driving, 3D model reconstruction, and object detection.  ...  To predict accurate disparity map, we propose a novel deep learning architecture for detectingthe disparity map from a rectified pair of stereo images, called MSDC-Net.  ...  It shows that multiple neural network architectures specifically adapt to the stereo matching task [25] .  ... 
arXiv:1904.12658v2 fatcat:3ohw3qmkvrbvthpcuis2biqe3m

Fast Robust Monocular Depth Estimation for Obstacle Detection with Fully Convolutional Networks [article]

Michele Mancini, Gabriele Costante, Paolo Valigi, Thomas A.Ciarfuglia
2016 arXiv   pre-print
We achieve these results using a Deep Neural Network approach trained on real and synthetic images and trading some depth accuracy for fast, robust and consistent operation.  ...  Obstacle Detection is a central problem for any robotic system, and critical for autonomous systems that travel at high speeds in unpredictable environment.  ...  The most trivial solutions are based on finding disparities between the two matched images, compute point clouds and apply heuristics to detect obstacles.  ... 
arXiv:1607.06349v1 fatcat:kshaumgb5feh7ko225vf44ljzu

Multi-scale Cross-form Pyramid Network for Stereo Matching [article]

Zhidong Zhu and Mingyi He and Yuchao Dai and Zhibo Rao and Bo Li
2019 arXiv   pre-print
We propose a novel deep learning architecture, which called CFP-Net, a Cross-Form Pyramid stereo matching network for regressing disparity from a rectified pair of stereo images.  ...  Stereo matching plays an indispensable part in autonomous driving, robotics and 3D scene reconstruction.  ...  The key point of this work is for stereo matching with a better matching cost.  ... 
arXiv:1904.11309v3 fatcat:3odjr6vc7rgafpp3nxgx5y526m

UNCERTAINTY ESTIMATION FOR END-TO-END LEARNED DENSE STEREO MATCHING VIA PROBABILISTIC DEEP LEARNING

M. Mehltretter
2020 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
Based on the well-known and commonly employed GC-Net architecture, a novel probabilistic neural network is presented, for the task of joint depth and uncertainty estimation from epipolar rectified stereo  ...  For this purpose, in the present work the idea of probabilistic deep learning is applied to the task of dense stereo matching for the first time.  ...  the Research Training Group i.c.sens [GRK2159], the MOBILISE initiative of the Leibniz University Hannover and TU Braunschweig and by the NVIDIA Corporation with the donation of the Titan V GPU used for  ... 
doi:10.5194/isprs-annals-v-2-2020-161-2020 fatcat:ks32kkzo3zgzfprfne27xhmdze

On the Synergies between Machine Learning and Binocular Stereo for Depth Estimation from Images: a Survey [article]

Matteo Poggi, Fabio Tosi, Konstantinos Batsos, Philippos Mordohai, Stefano Mattoccia
2021 arXiv   pre-print
While machine, and especially deep, learning advanced the state-of-the-art in stereo matching, stereo itself enabled new ground-breaking methodologies such as self-supervised monocular depth estimation  ...  based on deep networks.  ...  classifier, as ground control points (GCPs).  ... 
arXiv:2004.08566v2 fatcat:wcwfgzibo5evbkun3atpsz6kwm

Uncertainty Estimation for End-To-End Learned Dense Stereo Matching via Probabilistic Deep Learning [article]

Max Mehltretter
2020 arXiv   pre-print
Based on the well-known and commonly employed GC-Net architecture, a novel probabilistic neural network is presented, for the task of joint depth and uncertainty estimation from epipolar rectified stereo  ...  For this purpose, in the present work the idea of probabilistic deep learning is applied to the task of dense stereo matching for the first time.  ...  the Research Training Group i.c.sens [GRK2159], the MOBILISE initiative of the Leibniz University Hannover and TU Braunschweig and by the NVIDIA Corporation with the donation of the Titan V GPU used for  ... 
arXiv:2002.03663v1 fatcat:ne44thf53fguzhot7lznnrzc4m

3D Reconstruction of Curvilinear Structures with Stereo Matching DeepConvolutional Neural Networks [article]

Okan Altingövde, Anastasiia Mishchuk, Gulnaz Ganeeva, Emad Oveisi, Cecile Hebert, Pascal Fua
2021 arXiv   pre-print
We propose a fully automated pipeline for both detection and matching of curvilinear structures in stereo pairs by utilizing deep convolutional neural networks (CNNs) without making any prior assumption  ...  In this work, we mainly focus on 3D reconstruction of dislocations from stereo pairs of TEM images.  ...  Marc Legros for providing the HEA sample, Dr. Duncan T.L. Alexander for fruitful discussions, and Daniele Laub for help with TEM sample preparation.  ... 
arXiv:2110.07766v1 fatcat:agckxljcqvc5zd2dteoy2xvocq

Edge-preserving Stereo Matching Using Minimum Spanning Tree

Congxuan Zhang, Chao He, Zhen Chen, Wen Liu, Ming Li, Junjie Wu
2019 IEEE Access  
INDEX TERMS Stereo matching, minimum spanning tree, edge-preserving, brightness information, disparity range estimation, segmentation optimization.  ...  In this paper, we propose a minimum spanning tree (MST) based stereo matching method by using the image edge and segmentation optimization to preserve the image boundary.  ...  For instance, Bontar and Lecun [37] employed convolutional neural networks to learn similarity measures and proposed a new computational method for matching cost.  ... 
doi:10.1109/access.2019.2958527 fatcat:nhz5zexperdpnpc5pumy2jztrq

Front Matter: Volume 10696

Jianhong Zhou, Petia Radeva, Dmitry Nikolaev, Antanas Verikas
2018 Tenth International Conference on Machine Vision (ICMV 2017)  
The publisher is not responsible for the validity of the information or for any outcomes resulting from reliance thereon.  ...  Please use the following format to cite material from these proceedings: Publication of record for individual papers is online in the SPIE Digital Library.  ...  video stream recognition systems [10696-39] 10696 1V Recurrent neural network based virtual detection line [10696-48] 10696 1W First stereo video dataset with ground truth for remote car pose estimation  ... 
doi:10.1117/12.2319685 dblp:conf/icmv/X17 fatcat:jb7w6d2ewbe57gaqkvxcumxpum

MarkerPose: Robust Real-time Planar Target Tracking for Accurate Stereo Pose Estimation [article]

Jhacson Meza, Lenny A. Romero, Andres G. Marrugo
2021 arXiv   pre-print
MarkerPose is meant for high-accuracy pose estimation applications. Our method consists of two deep neural networks for marker point detection.  ...  The marker's pose is estimated through stereo triangulation. The target point detection is robust to low lighting and motion blur conditions.  ...  Meza thanks Universidad Tecnologica de Bolivar for a post-graduate scholarship and MinCiencias, and MinSalud for a "Joven Talento" scholarship.  ... 
arXiv:2105.00368v2 fatcat:dwlhqrx4drf3xfeyd56nizasoe

Autonomous Driving with Deep Learning: A Survey of State-of-Art Technologies [article]

Yu Huang, Yue Chen
2020 arXiv   pre-print
We investigate the major fields of self-driving systems, such as perception, mapping and localization, prediction, planning and control, simulation, V2X and safety etc.  ...  Due to the limited space, we focus the analysis on several key areas, i.e. 2D and 3D object detection in perception, depth estimation from cameras, multiple sensor fusion on the data, feature and task  ...  In supervised learning domain, there are different deep leaning methods, including Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN) as Long Short-Term Memory  ... 
arXiv:2006.06091v3 fatcat:nhdgivmtrzcarp463xzqvnxlwq
« Previous Showing results 1 — 15 out of 2,041 results