Accurate, Dense, and Robust Multi-View Stereopsis

By Yasutaka Furukawa, Jean Ponce. 2007

2,593 citations recorded by the source corpus.

Subfields: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Environmental Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

In Environmental Engineering

Citations in the corpus

SurfaceNet: An End-to-End 3D Neural Network for Multiview Stereopsis (2017, methodology): “The proposed method in [8] describes a patch model that consists of a quasi-dense set of rectangular patches covering the surface.”
Learning a Multi-View Stereo Machine (2017, background): “Another approach to MVS involves representing the reconstruction as a collection of depth maps [6, 57, 41, 13, 40].”
OctNetFusion: Learning Depth Fusion from Data (2017, background): “While image based techniques [38, 15, 12, 47, 37] provide compelling results for sufﬁciently textured surfaces, the introduction of affordable depth sensors, in particular the Microsoft Kinect sensor, allows scanning a wide variety of objects and scenes.”
Deep blending for free-viewpoint image-based rendering (2018, methodology): “Multi-view stereo algorithms (e.g., [Furukawa and Ponce 2010; Goe-sele et al. 2007; Jancosek and Pajdla 2011]) perform automatic 3D geometry reconstruction from unstructured photo datasets with diverse viewpoints.”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, positive): “The representative work for these kind of approaches is Patch-based Multi-View Stereo (PMVS) by Furukawa & Ponce (2010).”
Cost Volume Pyramid Based Depth Inference for Multi-View Stereo (2019, methodology): “ residual inference levels. Similar to previous methods [5,42,43], we use 5 views and apply the same depth map fusion method to obtain the point clouds. We Method Acc. Comp. Overall (mm) Geometic Furu[11] 0.613 0.941 0.777 Tola[39] 0.342 1.190 0.766 Camp[3] 0.835 0.554 0.695 Gipuma[12] 0.283 0.873 0”
Ray Markov Random Fields for image-based 3D modeling: Model and efficient inference (2010, background): “Quantitatively it is better than other volumetric methods, and competitive to the state-of-the-art feature based methods, e.g. [ 5].”
Semantic 3D Reconstruction with Continuous Regularization and Ray Potentials Using a Visibility Consistency Constraint (2016, neutral): “[8] 0.”
OpenMVG: Open Multiple View Geometry (2016, methodology): “While some solutions deliver a software program (Bundler [1], ColMap [39], MicMac [40], PMVS [35], CMVS [34]), others deliver both a collection of libraries and softwares (MVE [37], OpenMVG, OpenSfM [41], OpenMVS [42], TheiaSfM [38])10.”
CZECH TECHNICAL UNIVERSITY IN PRAGUE (2005, methodology): “Knowing camera poses, one can use multi-view stereo 3D surface reconstruction methods [9, 17] to construct compelling 3D surface models of the scene, see Figure 1.”
A dataset for developing and benchmarking active vision (2017, methodology): “To label every object instance in each scan, we feed the output of COLMAP into the dense reconstruction system CMVS/PMVS [28], [32].”
Street-view change detection with deconvolutional networks (2016, methodology): “In this section, we ﬁrst describe the suitability of our dense reconstruction method for the task of image alignment, compared to the output of [8] utilized by [38].”
ARCH: Animatable Reconstruction of Clothed Humans (2020, methodology): “Accurate methods based on stereo or fusion have been proposed using various types of sensors [12, 42, 31, 33, 38, 49, 50], and several applications have become popular in sports, medicine and entertainment (e.”
Learning to Infer Implicit Surfaces without 3D Supervision (2019, background): “However, 3D model databases are still limited compared to the massive availability of 2D photos, especially since acquiring clean and high-fidelity ground-truth 3D models still requires a tedious 3D capture process [24, 25].”
Few-Shot Generalization for Single-Image 3D Reconstruction via Priors (2019, background): “Classically, the problem of 3D reconstruction has been solved using multiple views and geometric or photometric constraints [6, 13].”