The Visual Hull Concept for Silhouette-Based Image Understanding

By Aldo Laurentini. 1994

Subfields: Computer Vision and Pattern Recognition, 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 Computer Vision and Pattern Recognition

In 3D Vision & Pose

Citations in the corpus

Multi-view Supervision for Single-View Reconstruction via Differentiable Ray Consistency (2017, background): “Visual hull [21] formalizes the notion of consistency between a 3D shape and observed object masks.”
3D Shape Induction from 2D Views of Multiple Objects (2016, methodology): “Visual-hull techniques [18] can be used to infer the shape of a particular object given its views from known viewpoints by taking the intersection of the projected silhouettes.”
Lifting Object Detection Datasets into 3D (2015, methodology): “After jointly estimating the camera viewpoints for all the instances in each class, we reconstruct the 3D shape of all objects using shape information borrowed from other exem-plars in the same class 4 .”
Category-specific object reconstruction from a single image (2014, methodology): “Two prominent approaches are visual hulls [24] and variational methods derived from snakes e.”
Background Subtraction in Real Applications: Challenges, Current Models and Future Directions (2019, methodology): “Practically, visual hull is a geometric entity obtained with shape-from-silhouette 3D reconstruction technique introduced by Laurentini [203].”
Consistent Depth Maps Recovery from a Video Sequence (2009, background): “Our experiments demonstrate that it is rather effective to improve the depth estimation around the occlusion boundaries and in the textureless regions.”
Vision-based hand pose estimation: A review (2007, methodology): “[120] used an approximate but fast reconstruction method based on the visual hull [121] that uses the silhouettes to compute a bounding volume for the object.”
Projective analysis for 3D shape segmentation (2013, background): “2004], where 2D data as sparse as object silhouettes can be highly effective [Laurentini 1994].”
Robust Probabilistic Occupancy Grid Estimation from Positive and Negative Distance Fields (2012, methodology): “Methods that operate on image inputs such as voxel coloring [29] or space carving [21] are considered out of scope in this discussion, as are silhouette-based methods [31, 22, 5].”
High Accuracy and Visibility-Consistent Dense Multiview Stereo (2012, unknown): “A number of multiview stereo algorithms have been proposed that exploit the visual hull [41].”
Space-time isosurface evolution for temporally coherent 3D reconstruction (2004, result): “The geometry reconstructed by this scheme is significantly better than results obtained by space-carving approaches that do not enforce temporal coherence.”
An Occupancy-Depth Generative Model of Multi-view Images (2007, background): “In fact, geometric occlusions are so important that there are reconstruction techniques, like those based on the visual hull, that use only the occlusion information [4,5].”
A Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms (2006, methodology): “hull [46] that serves as an initial estimate of scene geometry [5, 19, 31, 47, 48].”
Neural 3D Video Synthesis from Multi-view Video (2021, unknown): “Multi-view stereo [15, 49] and visual hull reconstructions [13, 27] have been successfully employed.”
A Review of the Evolution of Vision-Based Motion Analysis and the Integration of Advanced Computer Vision Methods Towards Developing a Markerless System (2018, background): “…the neutrally posed mesh, and then the mesh can be rotated around these joints to produce a posed body (right) of cameras are available, silhouettes can be combined into a 3D representation known as a visual hull [83], which is an approximation of the space occupied by the observed person (Fig.”
Methods for Volumetric Reconstruction of Visual Scenes (2004, unknown): “Perhaps the simplest class of volumetric multi-view reconstruction methods are visual hull approaches [25, 29, 49].”