Using Multiple Hypotheses to Improve Depth-Maps for Multi-View Stereo

By Neill D. F. Campbell, George Vogiatzis, Carlos Hernández, Roberto Cipolla. 2008

447 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

SurfaceNet: An End-to-End 3D Neural Network for Multiview Stereopsis (2017, methodology): “Figure 1: Reconstruction of the model 13 of the DTU dataset [1] in comparison to [4, 8, 27, 9].”
Multiview Neural Surface Reconstruction by Disentangling Geometry and Appearance (2020, methodology): “Assuming known cameras, classic Multi-View Stereo (MVS) methods [9, 40, 3, 45] try to reproduce the depth information by matching features points across views.”
Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness (2019, background): “A depth map represents dense 3D information that is perfectly aligned with a reference view; depth reconstruction has been demonstrated in many previous works on reconstruction with both single view [9, 51, 16, 17, 62] and multiple views [4, 48, 18, 14, 43, 57, 43].”
Pixelwise View Selection for Unstructured Multi-View Stereo (2016, methodology): “Stereo methods have advanced in terms of accuracy, completeness, scalability, and benchmarking – from the minimal stereo setup with two views [21–24] to multi-view methods [9,10,14,15,25–28].”
PGSR: Planar-Based Gaussian Splatting for Efficient and High-Fidelity Surface Reconstruction (2024, unknown): “Multi-View Photometric Consistency: Drawing inspiration from multi-view Stereo (MVS methods) [29], [34], [60], we employ photometric multi-view consistency constraints based on plane patches.”
Grounding Image Matching in 3D with MASt3R (2024, unknown): “RealEstate10K RRA@15 RTA@15 mAA(30) mAA(30) (a) Colmap+SG [21,72] 36 Camp [13] 0.835 0.554 0.695 Furu [31] 0.613 0.941 0.777 Tola [90] 0.342 1.190 0.766 Gipuma [32] 0.283 0.873 0.578 MVSNet [110] 0.396 0.527 0.462 CVP-MVSNet [109] 0.296 0.406 0.351 UCS-Net [17] 0.338 0.349 0.344 CER-MVS [55] 0.359…”
Multi-View Mesh Reconstruction with Neural Deferred Shading (2022, unknown): “[2, 6, 18, 19, 26, 68, 72, 74].”
Differentiable Volumetric Rendering: Learning Implicit 3D Representations Without 3D Supervision (2019, methodology): “mpare against classical approaches that have 3D meshes as output. To this end, we run screened Poisson surface reconstruction (sPSR) [34] on the output of the classical MVS algorithms Campbell et al. [9], Furukawa et al. [18], Tola et al. [78], and Colmap [67]. We ﬁnd that the results on the DTU ben”
State of the art in high density image matching (2014, background): “In Campbell et al. (2008) and Toldo et al. (2013), each pixel of an image is associated with a number of candidate depths.”
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching (2019, methodology): “Multi-View Stereo According to the comprehensive survey [5], works in traditional muti-view stereo can be roughly categorised into volumetric methods [13, 14, 18, 33], which estimate the relationship between each voxel and surfaces; point cloud based methods [6, 19], which directly process 3D points”
Cost Volume Pyramid Based Depth Inference for Multi-View Stereo (2019, unknown): “Overall ( mm ) G e o m e ti c 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 0.578 Colmap[33, 34] 0.400 0.664 0.532 L ea r n i ng SurfaceNet[17] 0.450 1.040 0.745 MVSNet[42] 0.396 0.527 0.462 P-MVSNet[26] 0.406 0.434 0.420 R-MVSNet[43] 0.383…”
Just Look at the Image: Viewpoint-Specific Surface Normal Prediction for Improved Multi-View Reconstruction (2016, background): “In terms of completeness, we are second best, narrowly behind [6], which however has a lot lower accuracy (61% higher error).”
Point-Based Multi-View Stereo Network (2019, unknown): “(mm) Comp. (mm) Overall (mm) Camp [2] 0.835 0.554 0.695 Furu [5] 0.613 0.941 0.777 Tola [25] 0.342 1.190 0.766 Gipuma [6] 0.283 0.873 0.578 SurfaceNet [12] 0.450”
A complete statistical inverse ray tracing approach to multi-view stereo (2011, background): “…color I r weighted by the visibility, i.e., The value of each voxel color can be computed by accumulating the messages received from all the rays that passes through the voxel: where M cr → ℓ xr [1] and M cr → ℓ xr [2] denote respectively the ﬁrst and second value of the 2-d vector M cr → ℓ xr .”
PatchmatchNet: Learned Multi-View Patchmatch Stereo (2020, unknown): “(mm) Comp.(mm) Overall(mm) Camp [4] 0.835 0.554 0.695 Furu [15] 0.613 0.941 0.777 Tola [35] 0.342 1.190 0.766 Gipuma [16] 0.283 0.873 0.578 SurfaceNet [20] 0.450 1.040 0.745 MVSNet [42] 0.396 0.527 0.462 R-MVSNet [43] 0.383 0.452 0.417 CIDER [39] 0.417 0.437 0.427 P-MVSNet [28] 0.406 0.434 0.420…”