Filters








13,204 Hits in 4.1 sec

The application of topological data analysis to human motion recognition

Marcin Żelawski, Tomasz Hachaj
2021 Technical Transactions  
The aim of this paper is to present and evaluate a novel method based on topological data analysis (TDA) for motion capture (kinematic) processing and human action recognition.  ...  The classification process does not require the use of machine learning and dynamical systems theory.  ...  We then discuss state-of-the-art action classification with topological data analysis.  ... 
doi:10.37705/techtrans/e2021011 fatcat:r6ulpwwrzfhljpvo4nhkzx4ryu

Persistent Homology of Attractors For Action Recognition [article]

Vinay Venkataraman, Karthikeyan Natesan Ramamurthy, Pavan Turaga
2016 arXiv   pre-print
In this paper, we propose a novel framework for dynamical analysis of human actions from 3D motion capture data using topological data analysis.  ...  We model human actions using the topological features of the attractor of the dynamical system.  ...  Experimental Results The proposed framework for topological data analysis for action representation was evaluated on the motion capture dataset [4] .  ... 
arXiv:1603.05310v1 fatcat:jiqfxspzancx7m3rqd7nvu4xnu

Topology Dictionary for 3D Video Understanding

T. Tung, T. Matsuyama
2012 IEEE Transactions on Pattern Analysis and Machine Intelligence  
The model relies on 1) topology description and classification using Reeb graphs, and 2) a Markov motion graph to represent topology change states.  ...  The model consists of a topology-based shape descriptor dictionary which can be generated from either extracted patterns or training sequences.  ...  The synthetic data set is created using 14 articulated character models, each of which animated using 28 motion capture sequences.  ... 
doi:10.1109/tpami.2011.258 pmid:22745004 fatcat:hn4vet5x7jcjxgijszcuxgxisu

On Time-Series Topological Data Analysis: New Data and Opportunities

Lee M. Seversky, Shelby Davis, Matthew Berger
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Using this approach, free-form paths. learning tasks such as classification can be performed by • MOCAP [8]: Motion capture data capturing multiple constructing kernels directly  ...  types of human motion. • Kitchen [8]: Motion capture data capturing multiple 3.  ... 
doi:10.1109/cvprw.2016.131 dblp:conf/cvpr/SeverskyDB16 fatcat:ponjmlkdhnak7loo32fqaugshm

Perturbation Robust Representations of Topological Persistence Diagrams [chapter]

Anirudh Som, Kowshik Thopalli, Karthikeyan Natesan Ramamurthy, Vinay Venkataraman, Ankita Shukla, Pavan Turaga
2018 Lecture Notes in Computer Science  
Topological methods for data analysis present opportunities for enforcing certain invariances of broad interest in computer vision, including view-point in activity analysis, articulation in shape analysis  ...  We explore the use of the proposed descriptor on three applications: 3D shape analysis, view-invariant activity analysis, and non-linear dynamical modeling.  ...  Introduction Over the years, tools from topological data analysis (TDA) have been used to characterize the invariant structure of data obtained from a noisy sampling of an underlying metric space [24]  ... 
doi:10.1007/978-3-030-01234-2_38 fatcat:c5bouc5iqfczfp3r3yhqr7vxum

Perturbation Robust Representations of Topological Persistence Diagrams [article]

Anirudh Som, Kowshik Thopalli, Karthikeyan Natesan Ramamurthy, Vinay Venkataraman, Ankita Shukla, Pavan Turaga
2018 arXiv   pre-print
Topological methods for data analysis present opportunities for enforcing certain invariances of broad interest in computer vision, including view-point in activity analysis, articulation in shape analysis  ...  We explore the use of the proposed descriptor on three applications: 3D shape analysis, view-invariant activity analysis, and non-linear dynamical modeling.  ...  Introduction Over the years, tools from topological data analysis (TDA) have been used to characterize the invariant structure of data obtained from a noisy sampling of This work was supported in part  ... 
arXiv:1807.10400v1 fatcat:3vfztcnho5bf7aicwl2ye4gqju

3D mesh video retrieval: A survey

Antonios Danelakis, Theoharis Theoharis, Ioannis Pratikakis
2012 2012 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video (3DTV-CON)  
They all involve retrieval procedures and, consequently, classification methods in order to identify similar actions/motions and facial expressions.  ...  The approaches are primarily categorized according to the 3D model representation that they use and their feature extraction and classification methods.  ...  The analysis of volumetric data, obtained from the processing of 3D video sequences after enclosing the 3D object into one or more cylinders, is a very common action/motion retrieval methodology.  ... 
doi:10.1109/3dtv.2012.6365464 dblp:conf/3dtv/DanelakisTP12 fatcat:jzcrxwnft5atbgosuvyqudqppy

LABELLING HUMAN KINEMATICS DATA USING CLASSIFICATION MODELS [article]

Yuan Shi, Nihir Chadderwala, Ujjwal Ratan
2022 medRxiv   pre-print
These data points are available in C3D format that contains numerical data transformed from 3D data captured from the sensors.  ...  In this paper, we will describe approaches to pre-process the kinematics data from its raw format and label the data points with missing markers using classification models.  ...  used to infer a variety of human actions.  ... 
doi:10.1101/2022.02.18.22271206 fatcat:q5a4hemldbfozpjduvxgsyo4xy

Time-Series Human Motion Analysis with Kernels Derived from Learned Switching Linear Dynamics

Taketoshi Mori, Masamichi Shimosaka, Tatsuya Harada, Tomomasa Sato
2005 Transactions of the Japanese society for artificial intelligence  
Summary In this paper, we propose a novel kernel computation algorithm between time-series human motion data for online action recognition.  ...  Another experimental result using kernel principal component analysis shows that the proposed kernel has excellent performance in extracting and separating different action categories, such as walking  ...  SLDs. 2 Classification Task Evaluation with Real Motion Capture Data As noted above, the marginalized BoV kernel has good property for time-series action classification when we give an ideal SLD to  ... 
doi:10.1527/tjsai.20.197 fatcat:uhcdqtijtbb2vjgnwsg3ivs5fy

Persistence Codebooks for Topological Data Analysis [article]

Bartosz Zielinski, Michal Lipinski, Mateusz Juda, Matthias Zeppelzauer, Pawel Dlotko
2019 arXiv   pre-print
Persistent homology (PH) is a rigorous mathematical theory that provides a robust descriptor of data in the form of persistence diagrams (PDs) which are 2D multisets of points.  ...  Persistence codebooks represent PDs in a convenient way for machine learning and statistical analysis and have a number of favorable practical and theoretical properties including 1-Wasserstein stability  ...  Acknowledgements The work from this article were supported by the National Science Centre, Poland under grant agreement no 2015/19/D/ST6/01215; 2016-2019.  ... 
arXiv:1802.04852v4 fatcat:x4auvgtqojfsrmh4dd66emvyb4

Video-based analysis of motion skills in simulation-based surgical training

Qiang Zhang, Lin Chen, Qiongjie Tian, Baoxin Li, Cees G. M. Snoek, Lyndon S. Kennedy, Reiner Creutzburg, David Akopian, Dietmar Wüller, Kevin J. Matherson, Todor G. Georgiev, Andrew Lumsdaine
2013 Multimedia Content and Mobile Devices  
Experiments were designed and performed to validate the proposed approach using realistic data captured from resident surgeons in local hospitals.  ...  We develop an automatic, video-based approach to analyzing the motion skills of a surgeon in simulation-based surgical training, where a surgical action is captured by multiple video cameras with little  ...  ACKNOWLEDGMENTS The work was supported in part by a grant (Grant No. 0904778) from the National Science Foundation.  ... 
doi:10.1117/12.2005177 fatcat:k24k4fqzsff2zg6cbq4pnyfblm

Robust classification of human actions from 3D data

Loc Huynh, Thanh Ho, Quang Tran, Thang Ba Dinh, Tien Dinh
2012 2012 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)  
This representation helps cope with noisy data while enhances both computational efficiency and flexibility. Also, we propose to use Hidden Markov Model (HMM) to recognize temporal motion patterns.  ...  We address the problem of classifying human actions using a single depth sensor camera.  ...  ACKNOWLEDGMENT This research is supported by research funding from Advanced Program in Computer Science, University of Science, Vietnam National University -Ho Chi Minh City.  ... 
doi:10.1109/isspit.2012.6621298 dblp:conf/isspit/HuynhHTDD12 fatcat:jyoguvhqjnfibf4wos5qbodpzu

Shape Distributions of Nonlinear Dynamical Systems for Video-Based Inference

Vinay Venkataraman, Pavan Turaga
2016 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We provide experimental validation through action and gesture recognition experiments on motion capture and Kinect datasets.  ...  The specific applications of interest in this paper are: 1) activity recognition using motion capture and RGBD sensors, 2) activity quality assessment for applications in stroke rehabilitation, and 3)  ...  Motion Capture Dataset In the first experiment, we evaluate the performance of the proposed framework using 3-dimensional motion capture sequences of body joints of subjects performing actions released  ... 
doi:10.1109/tpami.2016.2533388 pmid:27824585 fatcat:pxqg3svrrfhetnk2kh4ynaus7a

Representation, Analysis, and Recognition of 3D Humans

Stefano Berretti, Mohamed Daoudi, Pavan Turaga, Anup Basu
2018 ACM Transactions on Multimedia Computing, Communications, and Applications (TOMCCAP)  
Then, we focus on the analysis and recognition of 3D humans from 3D static and dynamic data, considering many applications for body and face.  ...  Spatial and temporal representations have been used in a variety of analysis and recognition tasks (see Section 3).  ...  Motion capture data were represented by a descriptor based on the shape of signal probability density function, and SVM was used for classification. Karg et al.  ... 
doi:10.1145/3182179 fatcat:ds55t4md2na2tibtyg4llerf3q

A Literature Review: Geometric Methods and Their Applications in Human-Related Analysis

Wenjuan Gong, Bin Zhang, Chaoqi Wang, Hanbing Yue, Chuantao Li, Linjie Xing, Yu Qiao, Weishan Zhang, Faming Gong
2019 Sensors  
This review presents a literature review on geometric concepts, geometric methods, and their applications in human-related analysis, e.g., human shape analysis, human pose analysis, and human action analysis  ...  ., geometric features, are widely used in computer graphics and computer vision problems.  ...  Topological Data Analysis In a broader perspective, topological data analysis (TDA) is an approach to analyzing data using topological methods [77, 78] and is closely related to persistent homology,  ... 
doi:10.3390/s19122809 fatcat:nrnq4vrj7bfv7iv4aem5g6k4zu
« Previous Showing results 1 — 15 out of 13,204 results