2020
DOI: 10.1155/2020/3096858
|View full text |Cite
|
Sign up to set email alerts
|

Enhanced Human Action Recognition Using Fusion of Skeletal Joint Dynamics and Structural Features

Abstract: In this research work, we propose a method for human action recognition based on the combination of structural and temporal features. The pose sequence in the video is considered to identify the action type. The structural variation features are obtained by detecting the angle made between the joints during the action, where the angle binning is performed using multiple thresholds. The displacement vector of joint locations is used to compute the temporal features. The structural variation features and the tem… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
6
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
7
2

Relationship

0
9

Authors

Journals

citations
Cited by 12 publications
(6 citation statements)
references
References 40 publications
0
6
0
Order By: Relevance
“…In this chapter, 14 angles in the human skeleton are taken, and each angle is calculated by two bone vectors. As the recognition of complex human actions depends on the characteristic parameter information of these actions, relevant characteristic parameters should be selected or constructed according to the specific properties of the object to be identified [ 25 , 26 ]. In complex motion feature extraction, motion feature extraction can be divided into nonmodel and model-based methods based on whether human body model is needed or not.…”
Section: Recognition Methods Of Wushu Human Complex Movement Based On...mentioning
confidence: 99%
“…In this chapter, 14 angles in the human skeleton are taken, and each angle is calculated by two bone vectors. As the recognition of complex human actions depends on the characteristic parameter information of these actions, relevant characteristic parameters should be selected or constructed according to the specific properties of the object to be identified [ 25 , 26 ]. In complex motion feature extraction, motion feature extraction can be divided into nonmodel and model-based methods based on whether human body model is needed or not.…”
Section: Recognition Methods Of Wushu Human Complex Movement Based On...mentioning
confidence: 99%
“…To speed up convergence, a batch normalization layer (BN) was added after the global average pooling (GAP) layer. In [10], the model predicts the highest k labels for every untrimmed video by analyzing global video-level features. First, frame-level binary classification is combined with dynamic programming to come up with temporally trimmed activity proposals.…”
Section: Related Workmentioning
confidence: 99%
“…This module's objective is to produce a confidence map for boundary-matching [25,26]. The confidence scores for densely distributed proposals include the boundary-matching layer and a sequence of the convolutional layers 3D and 2D are included in the module for proposal evaluation.…”
Section: Module For Proposal Evaluationmentioning
confidence: 99%
“…Joints MOCAP features are used to track the movements of human joints because joints are the most significant parts involved in human movements [78]. We represent the skeleton as S = {J k |k = 1, 2, .…”
Section: Joints Mocapmentioning
confidence: 99%