A Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms

By S. M. Seitz, Brian Curless, James Diebel, Daniel Scharstein, Richard Szeliski. 2006

2,894 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Microsoft COCO: Common Objects in Context (2014, background): “Some notable datasets include the Middlebury datasets for stereo vision [20], multi-view stereo [36] and optical ﬂow [21].”
DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (2019, background): “Mesh accuracy, as defined in [47], is the minimum distance d such that 90% of generated points are within d of the ground truth mesh.”
RayNet: Learning Volumetric 3D Reconstruction with Ray Potentials (2018, background): “As a thorough survey on 3D reconstruction techniques is beyond the scope of this paper, we discuss only the most related approaches and refer to [7, 13, 29] for a more thorough review.”
SurfaceNet: An End-to-End 3D Neural Network for Multiview Stereopsis (2017, methodology): “We also applied our model to an object of the Middlebury MVS dataset [21].”
Learning a Multi-View Stereo Machine (2017, background): “Modern approaches focus on acquiring the full 3D geometry in the form of volumetric representations or polygonal meshes [48].”
OctNetFusion: Learning Depth Fusion from Data (2017, background): “While image based techniques [40, 15, 12, 49, 39] 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.”
Lifting Object Detection Datasets into 3D (2015, background): “Visual hull reconstruction is a frequent first step in multi-view stereo [56], provid-”
Taskonomy: Disentangling Task Transfer Learning (2018, background): “grained classification [44] and object recognition [23], multi-view reconstruction [36] and structure from motion [27], wide baseline matching [41], SLAM [35] and visual odometry[52].”
Cognitive Mapping and Planning for Visual Navigation (2017, background): “Mapping has also been the focus of a large body of work in computer vision using RGB and RGB-D sensors [30, 33, 47, 57].”
Playing for Benchmarks (2017, background): “The Middlebury benchmarks for stereo, multi-view stereo, and optical ﬂow played a pivotal role in the development of low-level vision algorithms [4, 57, 58, 59].”
MOT16: A Benchmark for Multi-Object Tracking (2016, background): “…past, the computer vision community has developed centralized benchmarks for numerous tasks including object detection [14], pedestrian detection [12], 3D reconstruction [39], optical ﬂow [7], [19], visual odometry [19], single-object short-term tracking [26], and stereo estimation [19], [37].”
Discrete optimization of ray potentials for semantic 3D reconstruction (2015, neutral): “An overview is given in [21].”
Detailed Real-Time Urban 3D Reconstruction from Video (2007, background): “Completeness measures how much of the building was reconstructed and is defined in a way similar to [55].”
Let There Be Color! Large-Scale Texturing of 3D Reconstructions (2014, background): “So far, texture acquisition has not attracted nearly as much attention as geometry acquisition: Current benchmarks such as the Middlebury multi-view stereo benchmark [17] focus only on geometry and ignore appearance aspects.”
Large-Scale Data for Multiple-View Stereopsis (2016, methodology): “The evaluations of Seitz et al. (2006), Strecha et al. (2008) consider a small number of 3D scenes that are thought to be representative of real-world application domains for MVS.”
Fusion of depth maps with multiple scales (2011, methodology): “One source of depth maps are multi-view stereo (MVS) systems, which recently attained renewed interest [Seitz et al. 2006].”
On the Uncertainty of Self-Supervised Monocular Depth Estimation (2020, background): “Moreover, the ill-posed nature of depth-frommono perception task makes this eventuality much more likely to occur compared to techniques leveraging scene geometry [59, 60].”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, methodology): “Taxonomies: Several categorizations of multi-view reconstruction algorithms have been proposed in the literature, typically considering the form of the photo-consistency function, the scene representation, visibility computation, priors, and initialization requirements as in Seitz et al. (2006). Fro”
Real-Time 3D Reconstruction in Dynamic Scenes Using Point-Based Fusion (2013, background): “[19]), or with active sensors, such as laser scanners or depth cameras, based on triangulation or timeof-flight (ToF) techniques.”