A Multi-view Stereo Benchmark with High-Resolution Images and Multi-camera Videos

By Thomas Schöps, Johannes L. Schönberger, Silvano Galliani, Torsten Sattler, Konrad Schindler, Marc Pollefeys, Andreas Geiger. Computer Vision and Pattern Recognition, 2017

1,187 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/1bd57b1eb2467d14461cbf5952d2d06ade81f277

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Learning to Recover 3D Scene Shape from a Single Image (2020, methodology): “To evaluate the generalizability of our proposed depth prediction method, we take 9 datasets which are unseen during training, including YouTube3D [6], OASIS [7], NYU [34], KITTI [14], ScanNet [8], DIODE [36], ETH3D [33], Sintel [4], and iBims-1 [22].”
RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching (2021, methodology): “It ranks first on the Middlebury leaderboard [28] and outperforms all published work on the ETH3D leaderboard [29].”
HITNet: Hierarchical Iterative Tile Refinement Network for Real-time Stereo Matching (2020, methodology): “ETH3D two view stereo.”
Learned Multi-patch Similarity (2017, background): “A host of rather successful multi-view stereo methods exist (see benchmark results such as [12, 18, 19, 22]).”
BAD SLAM: Bundle Adjusted Direct RGB-D SLAM (2019, methodology): “This approach is similar in principle to the pose refinement in [55].”
Kimera: From SLAM to spatial perception with 3D dynamic scene graphs (2021, background): “R O ] 2 4 Ja n 20 21 geometric reconstruction (e.g., SLAM (Cadena et al. 2016), Structure from Motion (Enqvist et al. 2011), and MultiView Stereo (Schöps et al. 2017)) and deep-learning-based semantic segmentation (e.g., (Garcia-Garcia et al. 2017; Krizhevsky et al. 2012; Redmon and Farhadi 2017; Re”
Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping (2019, background): “, SLAM [1], Structure from Motion [2], and Multi-View Stereo [3]) and deep-learning-based semantic”
DIODE: A Dense Indoor and Outdoor DEpth Dataset (2019, methodology): “More recently, the ETH3D dataset [29] is similar to DIODE in terms of sensing modality and diversity.”
UniDepth: Universal Monocular Metric Depth Estimation (2024, methodology): “More precisely, each method is tested on validation splits from SUN-RGBD [38] without NYU split, Diode Indoor [40] , IBims-1 [22], VOID [44] HAM-MER [21], ETH-3D [37], nuScenes [4], and DDAD [19] with split proposed in [34] and evaluated with official masks.”
VGGT: Visual Geometry Grounded Transformer (2025, neutral): “We also compare the accuracy of our predicted point cloud to DUSt3R and MASt3R on the ETH3D [65] dataset.”
CFNet: Cascade and Fused Cost Volume for Robust Stereo Matching (2021, background): “3, the disparity range of half-resolution images in Middlebury [22] is even more than 6 times larger than full-resolution images in ETH3D [24] (400 vs 64).”
GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Image (2024, methodology): “We assess our model’s efficacy across six zero-shot relative depth benchmarks, including NYUv2 [60], KITTI [13], ETH3D [55], Scan-Net [7], DIODE [64], and OmniObject3D [66].”
Hierarchical Deep Stereo Matching on High-Resolution Images (2019, methodology): “Sceneflow contains around 30k synthetic image pairs with dense disparity labels.”
ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes (2023, methodology): “Additionally, we refine the camera poses with dense photometric error guided by the geometry of the laser scan [55, 37].”
PatchmatchNet: Learned Multi-View Patchmatch Stereo (2020, methodology): “[24] and ETH3D [32].”
Deep Marching Cubes: Learning Explicit Surface Representations (2018, background): “However, the recent success of deep learning [19,20,38] and the availability of large 3D datasets [5, 6, 9, 26, 37] nourishes hope for models that are able to learn powerful 3D shape representations from data, allowing reconstruction even in the presence of missing, noisy and incomplete observations”
Metric3D v2: A Versatile Monocular Geometric Foundation Model for Zero-Shot Metric Depth and Surface Normal Estimation (2024, neutral): “12, including comic-style (Row 2) and CG(computer graphics)-generated objects (Row5)”
Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation (2023, neutral): “ETH3D [45] and DIODE [50] are two high-resolution datasets, both featuring depth maps derived from LiDAR sensor measurements.”