Filters








21,423 Hits in 5.1 sec

Differentiable Diffusion for Dense Depth Estimation from Multi-view Images [article]

Numair Khan, Min H. Kim, James Tompkin
2021 arXiv   pre-print
We present a method to estimate dense depth by optimizing a sparse set of points such that their diffusion into a depth map minimizes a multi-view reprojection error from RGB supervision.  ...  Then, we apply this to light field and wider baseline images via self supervision, and show improvements in both average and outlier error for depth maps diffused from inaccurate sparse points.  ...  Depth via Differentiable Diffusion Given a set of n multi-view images I ={I 0 ,I 1 ,...  ... 
arXiv:2106.08917v2 fatcat:if4dgfvvgvbenhby7n22ga3n4m

Shape and Reflectance Reconstruction in Uncontrolled Environments by Differentiable Rendering [article]

Rui Li, Guangmin Zang, Miao Qi, Wolfgang Heidrich
2022 arXiv   pre-print
In this paper, we propose an efficient method to reconstruct the scene's 3D geometry and reflectance from multi-view photography using conventional hand-held cameras.  ...  With the optimal scene parameters evaluated, photo-realistic novel views for various viewing angles and distances can then be generated by our approach.  ...  A Multi-View Stereo (MVS) system [29] is introduced for robust and efficient dense modeling from unstructured image collections and jointly estimate depth and surface normal. 3D reconstruction. 3D reconstruction  ... 
arXiv:2110.12975v2 fatcat:uosmgsv5ivhbbkuejz44xsn2oa

Dense Multi-view 3D-reconstruction Without Dense Correspondences [article]

Yvain Quéau, Jean Mélou, Jean-Denis Durou, Daniel Cremers
2017 arXiv   pre-print
We introduce a variational method for multi-view shape-from-shading under natural illumination.  ...  The key idea is to couple PDE-based solutions for single-image based shape-from-shading problems across multiple images and multiple color channels by means of a variational formulation.  ...  Introduction Multi-view Shape-from-shading Over the decades the reconstruction of dense 3D geometry from images has been tackled in numerous ways.  ... 
arXiv:1704.00337v1 fatcat:osojqrgkxnf5jn2bqlfil5k5yi

LIGHTING AND SHADOW INTERPOLATION USING INTRINSIC LUMIGRAPHS

YASUYUKI MATSUSHITA, STEPHEN LIN, HEUNG-YEUNG SHUM, XIN TONG, SING BING KANG
2004 International Journal of Image and Graphics  
We decompose light fields captured at different lighting conditions into intrinsic images (reflectance and illumination images), and estimate view-dependent scene geometries using multi-view stereo.  ...  In this paper, we propose an approach to image-based lighting interpolation that is based on estimates of geometry and shading from relatively few images.  ...  From a sequence of N light fields, for each reference view we first obtain N estimated depth maps D(n) and N confidence maps C(n) using the multi-view stereo algorithm.  ... 
doi:10.1142/s0219467804001555 fatcat:ps77nzwagzfwxagfnrpcjzjjg4

Relightable Buildings from Images [article]

F. Melendez, M. Glencross, G. J. Ward, R. J. Hubbold
2011 Eurographics State of the Art Reports  
Exemplars of materials are obtained through surface depth hallucination, and our novel method matches these with multi-view image sequences that are also used to automatically recover 3D geometry.  ...  The requirements of our approach are that image capture must be performed under diffuse lighting and surfaces in the images must be predominantly Lambertian.  ...  Using the reconstructed model and camera parameters we create a high-resolution texture mosaic for the complete model from the multi-view image sequence by selecting the best view for each texel.  ... 
doi:10.2312/eg2011/areas/033-040 fatcat:7xpm2zjnxvgpndtoj4a5bis7fm

Deep 3D Capture: Geometry and Reflectance from Sparse Multi-View Images [article]

Sai Bi, Zexiang Xu, Kalyan Sunkavalli, David Kriegman, Ravi Ramamoorthi
2020 arXiv   pre-print
We propose a novel multi-view reflectance estimation network architecture that is trained to pool features from these coarsely aligned images and predict per-view spatially-varying diffuse albedo, surface  ...  We first estimate per-view depth maps using a deep multi-view stereo network; these depth maps are used to coarsely align the different views.  ...  We estimate the depth for each input view using a deep multi-view stereo network [51, 54] (Sec. 3.1).  ... 
arXiv:2003.12642v2 fatcat:zeqpyghkcjhofkdri5nhif77ai

Multi-View Edge-based Stereo by Incorporating Spatial Coherence

Gang Li, Yakup Genc, Steven W. Zucker
2007 3-D Digital Imaging and Modeling  
Texture-less scenes with clutter are traditionally difficult for dense multi-view stereo methods.  ...  The method utilizes the geometric consistency derived from the continuity of three-dimensional edge points as a geometric constraint in addition to previously known multi-view image constraints.  ...  [23] use 3D curve inference for regularizing diffusion MRI data. Multi-view geometry [7, 11] confines the locations of a given scene point in multiple images.  ... 
doi:10.1109/3dim.2007.35 dblp:conf/3dim/LiGZ07 fatcat:wyw77skknzclvf2c4ab6sqt5o4

3D Web Reconstruction of a Fibrous Filter Using Sequential Multi-Focus Images

Lingjie Yu, Guanlin Wang, Chao Zhi, Bugao Xu
2019 CMES - Computer Modeling in Engineering & Sciences  
Fujii, H.; Kodama, K.; Hamamoto, T. (2016): Scene flow estimation through 3D analysis of multi-focus images.  ...  Then a fusion image was established by extracting fiber edges from each layered image.  ...  Based on a synthesis of multi-focus images from multi-view images, Fujii et al.  ... 
doi:10.32604/cmes.2019.04494 fatcat:khvytur5tzdvfnsmgeivci7i5q

3D Scene Reconstruction from Multiple Spherical Stereo Pairs

Hansung Kim, Adrian Hilton
2013 International Journal of Computer Vision  
For accurate surface reconstruction, we propose a PDE-based disparity estimation method which produces continuous depth fields with sharp depth discontinuities even in occluded and highly textured regions  ...  Reconstruction is based on stereo image pairs with a vertical displacement between camera views. A 3D mesh model for each pair of spherical images is reconstructed by stereo matching.  ...  This estimates the camera parameters, as well as dense depth maps for the uploaded images.  ... 
doi:10.1007/s11263-013-0616-1 fatcat:mbk53thqwzbvrjkkqh6pvx4pc4

Shape from Depth Discontinuities [chapter]

Gabriel Taubin, Daniel Crispell, Douglas Lanman, Peter Sibley, Yong Zhao
2009 Lecture Notes in Computer Science  
An image capture process detects points on a depth discontinuity sweep from a camera moving with respect to an object, or from a static camera and a moving object.  ...  Locally convex points deep inside concavities can be estimated from the visible non-silhouette depth discontinuity points.  ...  Additional primal space information, such as from triangulation-based sensors or multi-view stereo photometric information, is needed to differentiate amongst these shapes and to produce a more accurate  ... 
doi:10.1007/978-3-642-00826-9_9 fatcat:vkvar6lwordbzgim366p2hdosm

Beyond Silhouettes: Surface Reconstruction Using Multi-Flash Photography

Daniel Crispell, Douglas Lanman, Peter G. Sibley, Yong Zhao, Gabriel Taubin
2006 Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06)  
This paper introduces a novel method for surface reconstruction using the depth discontinuity information captured by a multi-flash camera while the object moves along a known trajectory.  ...  The method extends well-established differential and global shape-from-silhouette surface reconstruction techniques by incorporating the significant additional information encoded in the depth discontinuities  ...  Our system uses active illumination to estimate the depth discontinuities from image data.  ... 
doi:10.1109/3dpvt.2006.37 dblp:conf/3dpvt/CrispellLSZT06 fatcat:6wceq3cvefgdll2ntre5prz5se

Deep Learning based Monocular Depth Prediction: Datasets, Methods and Applications [article]

Qing Li, Jiasong Zhu, Jun Liu, Rui Cao, Qingquan Li, Sen Jia, Guoping Qiu
2020 arXiv   pre-print
Estimating depth from RGB images can facilitate many computer vision tasks, such as indoor localization, height estimation, and simultaneous localization and mapping (SLAM).  ...  In this survey, we first introduce the datasets for depth estimation, and then give a comprehensive introduction of the methods from three perspectives: supervised learning-based methods, unsupervised  ...  The resulting depth images preserve clear object boundaries. Another direction is to learn affinity from RGB for spatial diffusion for the sparse depth map.  ... 
arXiv:2011.04123v1 fatcat:by6swdegvvdrxk73ti46k2rj2e

CroMo: Cross-Modal Learning for Monocular Depth Estimation [article]

Yannick Verdié, Jifei Song, Barnabé Mas, Benjamin Busam, Aleš Leonardis, Steven McDonagh
2022 arXiv   pre-print
We propose a novel pipeline capable of estimating depth from monocular polarisation for which we evaluate various training signals.  ...  Learning-based depth estimation has witnessed recent progress in multiple directions; from self-supervision using monocular video to supervised methods offering highest accuracy.  ...  We propose a multi-modal training approach that allows for monocular depth estimation from polarisation images.  ... 
arXiv:2203.12485v2 fatcat:pcrpj4i5czcaxio6fhsaczmyrq

Dense depth map reconstruction: A minimization and regularization approach which preserves discontinuities [chapter]

Luc Robert, Rachid Deriche
1996 Lecture Notes in Computer Science  
We present a variational approach to dense stereo reconstruction which combines powerful tools such as regularization and multi-scale processing to estimate directly depth from a number of stereo images  ...  Area-based: In these approaches, dense depth maps are provided by correlating the grey levels of image patches in the views being considered, assuming that they present some similarity [19, 13] .  ...  Conclusion A variational approach to dense stereo reconstruction has been presented. It allows to estimate directly the depth from a number of stereo images, while preserving depth discontinuities.  ... 
doi:10.1007/bfb0015556 fatcat:lmxm2poccjb4vfo42c3roa7vue

Learning Depth with Convolutional Spatial Propagation Network [article]

Xinjing Cheng, Peng Wang, Ruigang Yang
2019 arXiv   pre-print
For the tasks of sparse to dense, a.k.a depth completion.  ...  We can append this module to any output from a state-of-the-art (SOTA) depth estimation networks to improve their performances.  ...  Single view depth estimation via CNN and CRF. We first review single view depth estimation since methods in this area motivated our design.  ... 
arXiv:1810.02695v3 fatcat:5cu5y23cungkfemw4xq3bflipa
« Previous Showing results 1 — 15 out of 21,423 results