Towards Probabilistic Volumetric Reconstruction Using Ray Potentials

By Ali O. Ulusoy, Andreas Geiger, Michael J. Black. 2015

63 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Environmental Engineering, Probabilistic & Bayesian Methods, 3D Vision & Pose

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

In Environmental Engineering

In Probabilistic & Bayesian Methods

In 3D Vision & Pose

Citations in the corpus

Occupancy Networks: Learning 3D Reconstruction in Function Space (2018, background): “Paschalidou et al. [52] introduced an architecture that predicts voxel occupancies from multiple images, exploiting multi-view geometry constraints [ 69].”
OctNet: Learning Deep 3D Representations at High Resolutions (2016, background): “Areas of application include depth fusion [23], image rendering [26] and 3D reconstruction [44].”
RayNet: Learning Volumetric 3D Reconstruction with Ray Potentials (2018, methodology): “Occlusion constraints are encoded using high-order ray potentials [19, 28, 35].”
Learning a Multi-View Stereo Machine (2017, background): “[39, 35, 54] directly reason about viewing rays in a voxel grid, while [34] recovers a quasi dense point cloud.”
Multi-view Supervision for Single-View Reconstruction via Differentiable Ray Consistency (2017, background): “[34] propose a probabilistic framework where marginal distributions can be efficiently computed.”
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.”
Image2Mesh: A Learning Framework for Single Image 3D Reconstruction (2017, background): “Several papers have proposed a 3D volumetric representation [3, 4, 10, 11, 20, 24, 27, 30, 33, 35, 38] so they can feed deep neural networks applying 3D convolutions, pooling, and other techniques that have been successfully applied to 2D images for the learning process.”
Semantic Multi-view Stereo: Jointly Estimating Objects and Voxels (2017, methodology): “Occlusions and textureless or reflective surfaces cause fundamental ambiguities in 3D reconstruction [4, 34].”
Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness (2019, background): “Ray tracing can be incorporated into this voxelized structure [49, 38, 50].”
Semantic 3D Reconstruction with Continuous Regularization and Ray Potentials Using a Visibility Consistency Constraint (2016, background): “This was formulated in [25, 9, 21, 22, 31] as a problem of finding a voxel labeling in terms of color and occupancy such that the first occupied voxel along a ray has a similar color as the pixel it projects to.”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, methodology): “More recent approaches use a probability map defined at regular voxel locations to encode the probability of occupancy (Bhotika et al. (2002); Pollard & Mundy (2007); Ulusoy et al. (2015)).”
MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction (2022, methodology): “Classic multi-view stereo (MVS) methods [2, 6–8, 35, 35, 62, 64, 65] consider either feature matching for depth estimation [6, 62] or represent shapes with voxels [2, 7, 8, 35, 51, 64, 72, 73].”
Neural volumes (2019, background): “Ulusoy et al. [2015], Savinov et al. [2016], and Paschalidou et al. [2018] formulate graph-based energy or inference objectives using ray potentials as constraints.”
UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction (2021, background): “We recognize this expression as the image formation model for solid objects [55] where the term o(xi) ∏”
Patches, Planes and Probabilities: A Non-Local Prior for Volumetric 3D Reconstruction (2016, methodology): “: Ray potentials penalize deviations of the model from the input images: References [1] Ulusoy, Geiger and Black: Towards Probabilistic Volumetric Reconstruction using Ray Potentials; 3DV 2015 - Best Paper Award [2] Pollard and Mundy: Change detection in a 3-d world; CVPR 2007 [3] Liu and Cooper: St”