Clustered Pose and Nonlinear Appearance Models for Human Pose Estimation

By Sam Johnson, Mark Everingham. 2010

912 citations recorded by the source corpus.

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

Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields (2016, methodology): “The spatial model for articulated pose is either based on tree-structured graphical models [10, 30, 4, 3, 26, 40, 17], which parametrically encode the spatial relationship between adjacent parts following a kinematic chain, or non-tree models [37, 31, 20, 18, 8] that augment the tree structure with…”
Convolutional Pose Machines (2016, background): “The classical approach to articulated pose estimation is the pictorial structures model [2, 3, 9, 14, 26, 27, 30, 43] in which spatial correlations between parts of the body are expressed as a tree-structured graphical model with kinematic priors that couple connected limbs.”
DensePose: Dense Human Pose Estimation in the Wild (2018, background): “Our approach can be understood as the next step in the line of works on human pose estimation [27, 1, 20, 7, 42, 19, 29].”
Minding the Gaps for Block Frank-Wolfe Optimization of Structured SVMs (2016, methodology): “We build our model based on the one by Kolesnikov et al. (2014) for HorseSeg, and the one by Chen & Yuille (2014) for LSP.”
DeepPose: Human Pose Estimation via Deep Neural Networks (2013, methodology): “The second dataset we use is Leeds Sports Dataset [12] and its extension [13], which we will jointly denote by LSP. Combined they contain 11000 training and 1000 testing images.”
Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation (2014, methodology): “We evaluated our architecture on the extended-FLIC [29] and extended-LSP [18] datasets.”
Stacked Hourglass Networks for Human Pose Estimation (2016, background): “Recent pose estimation systems [15–20] have universally adopted ConvNets as their main building block, largely replacing hand-crafted features and graphical models; this strategy has yielded drastic improvements on standard benchmarks [1,21,22].”
Detect What You Can: Detecting and Representing Objects Using Holistic Models and Body Parts (2014, methodology): “The recent works on human pose estimation [25, 16, 22, 20] typically do not consider the holistic object (root) template in their model as they are faced with highly variable poses.”
Recent advances in convolutional neural networks (2015, methodology): “From the experiments, DeepPose can outperform the deformable part models based methods [149–151] on two widely used datasets, FLIC [152] and LSP [153].”
Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “Besides manual human skeletal joint annotation [56], [80], [81], a number of approaches have been designed to automatically construct a skeleton model from perception data.”
End-to-End Recovery of Human Shape and Pose (2017, methodology): “Note that LSP contains complex poses of people playing sports and no ground truth 3D labels are available for training.”
Efficient object localization using Convolutional Networks (2014, methodology): “In contrast to our work, they formulate the problem as a direct (continuous) regression to joint location rather than a discrete heat-map output.”
Human Pose Estimation with Iterative Error Feedback (2015, methodology): “We tested our method on the two most challenging benchmarks for 2D human pose estimation: the MPII Human Pose dataset [2], which features significant scale variation, occlusion, and multiple people interacting, and Leeds Sports Pose dataset (LSP) [19] which features complex poses of people in sports”
Articulated Pose Estimation by a Graphical Model with Image Dependent Pairwise Relations (2014, methodology): “Now we consider the problem of learning the model parameters from images with labeled part locations, which is the data available in most of the human pose datasets [17, 7, 10, 20].”
Articulated Human Detection with Flexible Mixtures of Parts (2013, neutral): “9 Johnson [50] 85.”
DeeperCut: A Deeper, Stronger, and Faster Multi-person Pose Estimation Model (2016, methodology): “While early work used classiﬁers such as SVMs and AdaBoost [1,5,6,7], modern approaches build on diﬀerent ﬂavors of deep learning-based architectures [8,9,10,11].”
Viewpoints and keypoints (2014, methodology): “This problem has been particularly well studied for humans - tracing back from classic model-based approaches for video [27, 16] to more recent pictorial structure based approaches [34] on challenging single image based real world datasets like LSP[20] or MPII Human Pose [1].”
Strong Appearance and Expressive Spatial Models for Human Pose Estimation (2013, methodology): “Perhaps the most widespread representation is based on the discriminatively trained orientation invariant appearance templates [2, 26, 16, 23] composed of HOG [6] or shape context [19] features.”
Pose Machines: Articulated Pose Estimation via Inference Machines (2014, methodology): “Pictorial structure models [1,2,3,4,5,6], express the human body as a tree-structured graphical model with kinematic priors that couple connected limbs.”