Spacetime stereo: shape recovery for dynamic scenes

By Li Zhang, Brian Curless, S. M. Seitz. 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings., 2003

363 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/5110e63eb5afd28442cdac87c55d7970d0c93890

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

High-quality video view interpolation using a layered representation (2004, background): “Inaninterestingtwisttoconventionallocalwindowmatching,Zhang et al. [2003] use matching windows that straddle space and time.”
High-Resolution Stereo Datasets with Subpixel-Accurate Ground Truth (2014, background): “Establishing ground-truth disparities from the input views directly avoids the calibration problem and can be done via unstructured light [1, 6, 34], but only yields disparities for nonoccluded scene points visible in both input images.”
Fast 3D Scanning with Automatic Motion Compensation (2007, background): “The authors of [20] and [5] present recent systems combining structured light and stereo into a common framework.”
Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Video (2020, methodology): “These methods are based on light ﬁelds [35, 19, 4], multi-view stereo to capture dense depth maps [99], layered depth images [70], or representations us-ing 3D point clouds [1, 40, 67], meshes [44, 86] or surfels [62, 7, 88] for dynamic scenes.”
Efficiently combining positions and normals for precise 3D geometry (2005, methodology): “To acquire positions, we use a temporal stereo triangulation scanner [Zhang et al. 2003; Davis et al. 2005], which combines the accuracy of active triangulation with the ease of calibration of stereo.”
Unique shape context for 3d data description (2010, methodology): “The other dataset, referred to as Lab, consists of scenes and models acquired in our laboratory by means of a 3D sensing technique known as Spacetime Stereo [6], [21].”
High-resolution, real-time three-dimensional shape measurement (2006, neutral): “Recently, Zhang et al.(2) and Davis et al.”
Spacetime stereo: a unifying framework for depth from triangulation (2005, background): “have simultaneously developedmethods similar to ours, focusing on recovery of dynamic scenes rather than on constructing an organizing framework [33].”
Recent progresses on real-time 3D shape measurement using digital fringe projection techniques (2010, methodology): “Spacetime stereo is another technique that has potential for high speed 3D shape measurement [2] [3] [4].”
SHOT: Unique signatures of histograms for surface and texture description (2014, methodology): “In the evaluation proposed in [49], focused on object recognition in clutter and occlusion from CAD models, (a) (b) (c) (d) Figure 10: 3D Reconstruction from Spacetime Stereo views (first two rows) and Kinect views (last two rows): (a) initial set of views (b) coarse registration (c) global registra”
Spacetime faces: high resolution capture for modeling and animation (2004, methodology): “Consequently, [ 6 ] developed an approach to minimize Eq. 13.4 by searching for the best matching sheared window at each pixel independently.The resulting depth maps are both higher-resolution and more stable than those produced using standard stereo matching as shown in Figs.”
High-accuracy stereo depth maps using structured light (2003, background): “While doing our research, we became aware of concurrent work aimed at acquiring high-quality correspondences with active illumination that is applicable to both static and dynamic scenes [8, 24].”