Recovering 3D human body configurations using shape contexts

By Greg Mori, Jitendra Malik. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006

305 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Human-Computer Interaction

View Paper: https://www.semanticscholar.org/paper/f853c25d899b2127f97436f23e81ffe8ff0a4c14

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 Human-Computer Interaction

Citations in the corpus

Posebits for Monocular Human Pose Estimation (2014, background): “Faced with measurement noise, missing data, and ambiguity, extensive use of 3D pose data has been common, either to learn generative pose priors, or discriminative mappings from image features to 3D pose (e.g. see [3, 12, 13, 18]).”
Deep Multitask Architecture for Integrated 2D and 3D Human Sensing (2017, background): “There is a vast literature on monocular 3d human pose estimation, including the analysis of kinematic ambiguities associated with the 3d pose [42, 5], as well as generative and discriminative methods for learning and inference [35, 37, 4, 1, 26, 38, 14, 30].”
VNect (2017, unknown): “CCS Concepts: • Computing methodologies → Reconstruction ; Computational photography ; Image-based rendering ;”
Multi-source Deep Learning for Human Pose Estimation (2014, background): “Pose estimation is considered as holistic recognition in [14, 32, 33].”
Learning to Fuse 2D and 3D Image Cues for Monocular Body Pose Estimation (2016, unknown): “Early approaches falling into this category typically worked by extracting hand-crafted features and learning a mapping from these features to 3D poses [1, 8, 28, 29, 37, 56, 68], or by retrieving 3D poses from a database based on similarity with the 2D image evidence [18, 26, 39, 41, 42].”
Exploiting Temporal Information for 3D Human Pose Estimation (2017, methodology): “Earlier methods used to focus on extracting features, invariant to factors such as background scenes, lighting, and skin color from images and mapping them into 3D human pose [2,3,4,5].”
Sparseness Meets Deepness: 3D Human Pose Estimation from Monocular Video (2015, background): “Another strand of research has focused on methods that predict 3D poses by searching a database of exemplars [36, 27, 19] or via a discriminatively learned mapping from the image directly or image features to human joint locations [1, 34, 51, 16, 44].”
Robust Estimation of 3D Human Poses from a Single Image (2014, methodology): “For example, Mori et al. [11] match a test image to the stored exemplars using shape context descriptors, and transfer the matched 2D pose to the test image.”
Monocular 3D Human Pose Estimation in the Wild Using Improved CNN Supervision (2016, background): “3D pose can be inferred from 2D pose through geometric and statistical priors [37, 59].”
3D Human Pose Estimation from a Single Image via Distance Matrix Regression (2016, background): “The two steps can be iteratively reﬁned [38, 51, 49] or formulated independently [31, 26, 10, 39].”
Shape and Appearance Context Modeling (2007, methodology): “We propose to compute the shape labeled image with a procedure that is inspired by the idea of shape context [1, 16].”
Robust Object Tracking with Online Multiple Instance Learning (2011, unknown): “These methods are closely related to object detection—an area that has seen great progress in the last decade—and are referred to as " tracking-by-detection " or " tracking by repeated recognition " [31].”
Fusing 2D Uncertainty and 3D Cues for Monocular Body Pose Estimation (2016, unknown): “In the past, this usually involved inferring a 3D human pose either optimizing an energy function derived from image information, such as silhouettes [5, 13, 20, 21, 24, 29, 43, 53], feature trajectories [68] and 2D joint locations [3, 4, 19, 33, 45, 51, 62, 63], or 2D recognitionbased pose retrieva”
Multi-task Recurrent Neural Network for Immediacy Prediction (2015, methodology): “Both holistic models [28, 11, 19, 20, 33] and local models [31, 8, 39, 6, 9, 30, 24, 10, 26, 14, 29, 35, 25, 1, 21, 4] were used for estimating the pose of a single person.”
Multi-view 3D Human Pose Estimation in Complex Environment (2011, methodology): “Pose initialization using 2D generative models involves 2D pose recovery (Andriluka et al. 2009; Ferrari et al. 2009; Mori and Malik 2006; Ramanan et al. 2007) followed by a 3D inference step (Lee and Nevatia 2009) with respect to the joint locations.”
Synthesizing Training Images for Boosting Human 3D Pose Estimation (2016, background): “While some earlier works were based on local descriptors [34, 13], the recent emergence of CNNs has led to significant improvements in body pose estimation from a single image.”
Pose-conditioned joint angle limits for 3D human pose reconstruction (2015, background): “Discriminative approaches also exist in the literature that do not require 2D point correspondence and directly estimate human pose from 2D image measurements [3, 20, 22, 25, 34].”
MonoCap: Monocular Human Motion Capture using a CNN Coupled with a Geometric Prior (2017, background): “Another strand of research has focused on discriminative methods that predict 3D poses by searching a database of exemplars [29], [30], [31], [32] or via a discriminatively learned mapping from the image directly to human joint locations [33], [34], [35], [36], [37], [38].”
Monocular 3D pose estimation and tracking by detection (2010, background): “There is also a lot of work on predicting 3D poses directly from image features using regression [1] [16] [27], classification [21], or a search over a database of exemplars [19] [22].”