Filters








83,069 Hits in 7.2 sec

Image Matching across Wide Baselines: From Paper to Practice [article]

Yuhe Jin and Dmytro Mishkin and Anastasiia Mishchuk and Jiri Matas and Pascal Fua and Kwang Moo Yi and Eduard Trulls
2020 arXiv   pre-print
This work provides a basis for the Image Matching Challenge https://vision.uvic.ca/image-matching-challenge.  ...  Data and code are online https://github.com/vcg-uvic/image-matching-benchmark, providing an easy-to-use and flexible framework for the benchmarking of local features and robust estimation methods, both  ...  Despite decades of research, image matching remains unsolved in the general, wide-baseline scenario.  ... 
arXiv:2003.01587v3 fatcat:kf6lbmyiyrbrvce2oxsluv7xpq

Variable baseline/resolution stereo

David Gallup, Jan-Michael Frahm, Philippos Mordohai, Marc Pollefeys
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
Many datasets, such as video captured from a moving camera, allow the baseline to be selected with significant flexibility.  ...  We present a novel multi-baseline, multi-resolution stereo method, which varies the baseline and resolution proportionally to depth to obtain a reconstruction in which the depth error is constant.  ...  The baseline and resolution increase from narrow to wide and from coarse to fine as necessary to maintain the target error bound.  ... 
doi:10.1109/cvpr.2008.4587671 dblp:conf/cvpr/GallupFMP08 fatcat:g4arx7i52fcldezzumtrbi2pr4

Robust Quasi-Dense Wide Baseline Match Propagation

Yiran Wang, Wei Wang, Hai Zhu, Xiaowei Zhang
2016 ICIC Express Letters  
In order to overcome the limitations of traditional wide baseline match propagation methods, the proposed method simultaneously models illumination variations and perspective distortions in images, and  ...  The paper proposes a robust three-view matching method to produce quasidense 3D points in multi-view stereo.  ...  In the paper, we have proposed a robust wide baseline matching method which can be used to generate quasi-dense reliable 3D points in multi-view stereo.  ... 
doi:10.24507/icicel.10.11.2527 fatcat:ysw2msdroraktap3uau5ncqpee

Vehicle Re-Identification: an Efficient Baseline Using Triplet Embedding [article]

Ratnesh Kumar, Edwin Weill, Farzin Aghdasi, Parthsarathy Sriram
2019 arXiv   pre-print
Re-identification is the problem of matching appearances of objects across different cameras.  ...  In this paper we provide an extensive evaluation of these losses applied to vehicle re-identification and demonstrate that using the best practices for learning embeddings outperform most of the previous  ...  Table 13 Conclusion and Future Work In this paper we propose a strong baseline for vehicle re-identification using the best practices in learning deep triplet embedding [11] .  ... 
arXiv:1901.01015v4 fatcat:5d4u4aykyzfg5gnjh3nhpygidm

Edge descriptors for robust wide-baseline correspondence

Jason Meltzer, Stefano Soatto
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
This paper describes a method for finding wide-baseline correspondences between images at locations along gradient edges.  ...  We demonstrate the effectiveness of edge matching in the applications of wide-baseline correspondence, structure from motion from line segments, and object category recognition on the Caltech 101 dataset  ...  The authors would also like to thank Brian Fulkerson and Andrea Vedaldi for sharing code and ideas.  ... 
doi:10.1109/cvpr.2008.4587684 dblp:conf/cvpr/MeltzerS08 fatcat:57r652dhx5dczfvca4fglglj2e

Segmentation Based Features for Wide-Baseline Multi-view Reconstruction

Armin Mustafa, Hansung Kim, Evren Imre, Adrian Hilton
2015 2015 International Conference on 3D Vision  
wide-baseline views for images of a variety of indoor and outdoor scenes; and the number of wide-baseline matches is increased by an order of magnitude compared to alternative detector-descriptor combinations  ...  In this paper we introduce a novel segmentation based feature detector SFD that produces an increased number of 'good' features for accurate wide-baseline reconstruction.  ...  Contributions of this paper are: • A novel segmentation based feature detector SFD for accurate wide-baseline matching and sparse reconstruction from wide-baseline views; • SFD gives an increased number  ... 
doi:10.1109/3dv.2015.39 dblp:conf/3dim/MustafaKIH15 fatcat:vqwrxvlwzzhbroag3tcif3n2ku

Wide baseline correspondence extraction beyond local features

R. Lakemond, S. Sridharan, C. Fookes
2011 IET Computer Vision  
Robust, affine covariant, feature extractors provide a means to extract correspondences between images captured by widely separated cameras.  ...  Advances in wide baseline correspondence extraction require looking beyond the robust feature extraction and matching approach.  ...  Camera geometry may be computed automatically from sparse point correspondences across views. This is referred to as the wide baseline correspondence problem.  ... 
doi:10.1049/iet-cvi.2010.0045 fatcat:wwk4ki5tobaobohqazux4q2h2a

Detecting Text Baselines in Historical Documents with Baseline Primitives

Wei Jia, Chixiang Ma, Lei Sun, Qiang Huo
2021 IEEE Access  
primi-tive pairs from different text baselines as negative samples without the ambiguity associated with matching the learned baseline primitives to the ground-truth text baselines.  ...  Function analysis methods try to segment text lines by finding an optimal path across the document image. Saabni et al.  ... 
doi:10.1109/access.2021.3093568 fatcat:bzawhelpbrecvkszjgws2ag47m

ROVO: Robust Omnidirectional Visual Odometry for Wide-baseline Wide-FOV Camera Systems [article]

Hochang Seok, Jongwoo Lim
2019 arXiv   pre-print
In this paper we propose a robust visual odometry system for a wide-baseline camera rig with wide field-of-view (FOV) fisheye lenses, which provides full omnidirectional stereo observations of the environment  ...  The hybrid projection model combines the perspective and cylindrical projection to maximize the overlap between views and minimize the image distortion that degrades feature matching performance.  ...  In our wide-baseline wide-FOV setup there exist large overlapping area between images, but direct feature matching with the descriptors from such distorted areas does not yield good correspondences in  ... 
arXiv:1902.11154v2 fatcat:ijxu55ljfjhkxeo6oef75ueis4

Robust wide baseline pose estimation from video

Nicola Pellicano, Emanuel Aldea, Sylvie Le Hegarat-Mascle
2016 2016 23rd International Conference on Pattern Recognition (ICPR)  
Robust wide baseline pose estimation is an essential step in the deployment of smart camera networks.  ...  The results show a fast and robust convergence of the solution, and a significant improvement, compared to single image based alternatives, of the RMSE of ground truth matches, and of the maximum absolute  ...  ACKNOWLEDGMENT The authors gratefully acknowledge the support from Regent's Park Mosque for providing access to the site during data collection, and from K. Kiyani.  ... 
doi:10.1109/icpr.2016.7900230 dblp:conf/icpr/PellicanoAH16 fatcat:s5dk4knayrd45olrufvs27a6lq

Large-baseline matching and reconstruction from symmetry cells

Kun Huang, A.Y. Yang, Wei Hong, Yi Ma
2004 IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004  
In this paper, we study how the presence of symmetry in man-made environments may significantly facilitate the task of automatic matching features and recovering 3-D camera pose and scene structure from  ...  approach which allows automatic feature matching between images taken with arbitrary (both large and small) camera motions.  ...  Once a set of symmetry cells are extracted from each image, we still need to know how to match them across different views.  ... 
doi:10.1109/robot.2004.1308023 dblp:conf/icra/HuangYHM04 fatcat:eziemxvbs5edxb5n5rpiisme6q

Multi-baseline Disparity Fusion for Immersive Videoconferencing

Oliver Schreer, Nicole Atzpadin, Ingo Feldmann, Peter Kauff
2009 Proceedings of the 2nd International ICST Conference on Immersive Telecommunications  
These observations together with experimental investigations lead to a multi-baseline configuration that allows taking advantage of small and wide baseline stereo camera as well as trifocal camera configurations  ...  In this paper, we present a multibaseline disparity fusion scheme for improved real-time disparity map estimation.  ...  Figure 7 : 7 Outline of the pixel-recursion scheme Figure 8 -Figure 9 - 89 Original images: trifocal short baseline Original images: wide baseline Figure 10 - 10 Disparity images: Wide baseline (left  ... 
doi:10.4108/immerscom.2009.6585 dblp:conf/immerscom/SchreerAFK09 fatcat:rhkajyny6rhnnfamdlo54tamli

Dense Wide-Baseline Scene Flow From Two Handheld Video Cameras [article]

Christian Richardt, Hyeongwoo Kim, Levi Valgaerts, Christian Theobalt
2016 arXiv   pre-print
We propose a new technique for computing dense scene flow from two handheld videos with wide camera baselines and different photometric properties due to different sensors or camera settings like exposure  ...  First, we compute dense, wide-baseline correspondences using DAISY descriptors for matching between cameras and over time.  ...  We thus use insights from the design of DAISY and adapt it to the case of dense matching across camera views and over time.  ... 
arXiv:1609.05115v1 fatcat:pll2sifdqrftljxsx3gsrsl65m

Wide-baseline image matching using Line Signatures

Lu Wang, Ulrich Neumann, Suya You
2009 2009 IEEE 12th International Conference on Computer Vision  
We present a wide-baseline image matching approach based on line segments. Line segments are clustered into local groups according to spatial proximity.  ...  Similar to local features, line signatures are robust to occlusion, image clutter, and viewpoint changes. The descriptor and similarity measure of line signatures are presented.  ...  Fig.4 shows 2 image pairs from the line matching papers [2] and [16] .  ... 
doi:10.1109/iccv.2009.5459316 dblp:conf/iccv/WangNY09 fatcat:mbiwzwdqrrgkffpnd5pmu6ymja

Dense Wide-Baseline Scene Flow from Two Handheld Video Cameras

Christian Richardt, Hyeongwoo Kim, Levi Valgaerts, Christian Theobalt
2016 2016 Fourth International Conference on 3D Vision (3DV)  
We propose a new technique for computing dense scene flow from two handheld videos with wide camera baselines and different photometric properties due to different sensors or camera settings like exposure  ...  First, we compute dense, wide-baseline correspondences using DAISY descriptors for matching between cameras and over time.  ...  We thus use insights from the design of DAISY and adapt it to the case of dense matching across camera views and over time.  ... 
doi:10.1109/3dv.2016.36 dblp:conf/3dim/RichardtKVT16 fatcat:tbnfe27g45ervochzivraxjpem
« Previous Showing results 1 — 15 out of 83,069 results