Accurate 3D pose estimation from a single depth image

By Mao Ye, Xianwang Wang, Ruigang Yang, Liu Ren, Marc Pollefeys. 2011

272 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering

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 Control and Systems Engineering

Citations in the corpus

Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “[52] Motion exemplars Single depth image 3D skeleton, 38mm accuracy Jung et al.”
Human Action Recognition From Various Data Modalities: A Review (2020, unknown): “[167], [168] and depth data [169], [170], [171], [172], [173], [174].”
3D skeleton-based human action classification: A survey (2016, background): “Various attempts have been reported in literature to estimate body parts in RGB data [35,80,21,81] or in depth map [56,82].”
Articulated and Generalized Gaussian Kernel Correlation for Human Pose Estimation (2016, methodology): “We evaluate our articulated pose tracking algorithm on two depth benchmark datasets, i.e., [8] (body) and [19] (hand), which shows that the accuracy of pose estimation is competitive compared to the best results reported so far [6], [10], [20].”
A review of algorithms for filtering the 3D point cloud (2017, unknown): “To solve this problem, Ye et al. [53] projected only z coordinates of the point-set, then x and y could be calculated through re-projection.”
Random tree walk toward instantaneous 3D human pose estimation (2015, background): “Since the body part sizes are all different, an excessive number of tree evaluations are often made for a larger body part.”
Realtime and Robust Hand Tracking from Depth (2014, methodology): “Recent advances in human body tracking [24, 1, 15, 30, 27] have proven that the capability of re-initialization on every frame is critical for robust tracking.”
Towards Viewpoint Invariant 3D Human Pose Estimation (2016, methodology): “Dense point clouds provided by depth sensors motivate the use of iterative closest point algorithms [21,26,27,36] and database lookups [65].”
A Survey on Human Motion Analysis from Depth Data (2013, methodology): “There are mainly three ways to obtain skeletons: active motion capture (MoCap) systems, monocular or multi-view color images and single view depth maps [24, 25].”
Gesture recognition for human-robot collaboration: A review (2017, unknown): “Example of skeleton model identification (Ye et al., 2011).”
Monocular 3D Human Pose Estimation by Predicting Depth on Joints (2017, background): “ages [40, 26] or RGB images captured simultaneously at multiple viewpoints [2, 10].”
The Vitruvian manifold: Inferring dense correspondences for one-shot human pose estimation (2012, background): “Furthermore, the marginal improvements provided over the current state of the art [25] are bolstered by our algorithm running at least two orders of magnitude faster.”
Interactive Markerless Articulated Hand Motion Tracking Using RGB and Depth Data (2013, methodology): “Similar to previous work in full-body motion tracking [2, 26, 29] we instantiate two pose estimators in parallel.”
Gen: a general-purpose probabilistic programming system with programmable inference (2019, methodology): “We next consider a computer vision task [70].”