An Approach to Pose-Based Action Recognition

By Chunyu Wang, Yizhou Wang, Alan L. Yuille. 2013

394 citations recorded by the source corpus.

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

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Human-Computer Interaction

Citations in the corpus

Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “Bag-of-words encoding is widely employed by a large number of skeleton-based human representations [100], [106], [118], [134], [135], [138], [141], [143], [149], [152], [155], [157], [165], [170], [177], [179],”
3D Human Activity Recognition with Reconfigurable Convolutional Neural Networks (2014, background): “Several compositional approaches were studied for complex scenarios and achieved substantial progresses[33, 29, 39, 21, 23, 30, 4, 15], and they decomposed an activity into deformable parts and enriched the models with contextual information.”
Chained Multi-stream Networks Exploiting Pose, Motion, and Appearance for Action Classification and Detection (2017, methodology): “Wang et al. [39] used data mining techniques to obtain a representation for each video and ﬁnally, by us-ing a bag-of-words model to classify videos.”
Deep Learning on Lie Groups for Skeleton-Based Action Recognition (2016, background): “iority over existing shallow Lie group feature learning methods as well as most conventional deep learning methods. 1. Introduction Due to the development of depth sensors, 3D human activity analysis [27,45,23,43,41,3,42,37,44,26, 35,17] has attracted more interest than ever before. Recent manifold-”
A novel hierarchical framework for human action recognition (2016, methodology): “In most 3D skeleton-based methods [11, 12, 14, 16, 18, 21], mo-30 tion features are represented by pair-wise diﬀerences of joint positions within the current frame or between the current frame and the previous frames.”
Action Recognition Based on Joint Trajectory Maps Using Convolutional Neural Networks (2016, background): “oints by extracting spatial descriptor [30], [56]– poses [35], [61]–[63]. The groups of joints aim to detect the discriminative subsets of joints to differentiate actions. Methods such as [33], [61], [64]–[67] focus on mining the subsets of most discriminative joints or consider the correlation of p”
Articulated Pose Estimation by a Graphical Model with Image Dependent Pairwise Relations (2014, background): “Progress in this area can immediately be applied to important vision tasks such as human tracking [2], action recognition [25] and video analysis.”
A Hierarchical Pose-Based Approach to Complex Action Understanding Using Dictionaries of Actionlets and Motion Poselets (2016, background): “Similarly, recent works have also demonstrated the relevance of explicitly incorporating body pose information in action recognition models [11, 30].”
Enhanced Sequence Matching for Action Recognition from 3D Skeletal Data (2014, positive): “[10] group the joints into five body parts and uses data mining to obtain a spatial-temporal configurations of human actions.”
Efficient Pose-Based Action Recognition (2014, background): “While [21] weight poses of actions based on a mutual information criteria, [19] mine for most occurring temporal and spatial structures of body joints for classification.”
Mining Mid-Level Features for Action Recognition Based on Effective Skeleton Representation (2014, background): “[15] first estimated human joints positions from videos and then grouped the estimated joints into five parts.”
A Deep Structured Model with Radius–Margin Bound for 3D Human Activity Recognition (2015, background): “Several compositional part-based approaches have been studied for complex scenarios and achieved substantial progresses [46, 42, 54, 31, 33, 44, 6], and they represent an activity with the deformable parts and contextual relations.”
Kernelized covariance for action recognition (2016, background): “However, in general, the classification is subsequent to a design phase of discriminative features such as actionlets [8], random occupancy patterns [9], pose-based sets [10], space-time trajectories [11], velocity and acceleration [12], normal vectors [13] or Lie group geometry”
Predicting the Where and What of Actors and Actions through Online Action Localization (2016, background): “[33] also extended [40] to videos and represented videos in terms of spatio-temporal configurations of joints to perform action recognition.”
Robust Estimation of 3D Human Poses from a Single Image (2014, background): “Since an action is naturally represented by human poses [18], 2D and 3D pose estimation has attracted a lot of attention.”
Recognizing Actions in 3D Using Action-Snippets and Activated Simplices (2016, background): “A third line of works groups body joints into body parts by proposing a spatial-temporal-part model to capture the differentiable conﬁgurations of body parts for classiﬁcation (Wang, Wang, and Yuille 2013).”
Video Action Detection with Relational Dynamic-Poselets (2014, background): “Previous studies have shown that pose [23,13,10,17,7] and motion [27,30,15,5] are key elements in understanding human actions from videos.”