Filters








900 Hits in 5.0 sec

Object Properties Inferring from and Transfer for Human Interaction Motions [article]

Qian Zheng, Weikai Wu, Hanting Pan, Niloy Mitra, Daniel Cohen-Or, Hui Huang
2020 arXiv   pre-print
, leading to new synthesis possibilities for human interaction motions.  ...  Our results clearly demonstrate that the interaction motions and interacting objects are highly correlated and indeed relative object latent properties can be inferred from the 3D skeleton sequences alone  ...  Skeleton is also a special graph structure representation, and thus graph convolution networks are utilized as well for action recognition .  ... 
arXiv:2008.08999v1 fatcat:gnrrs7hfwraa5ncmayturw3zyi

Inferring object properties from human interaction and transferring them to new motions

Qian Zheng, Weikai Wu, Hanting Pan, Niloy Mitra, Daniel Cohen-Or, Hui Huang
2021 Computational Visual Media  
possibilities for motions involving human interaction.  ...  Our results clearly demonstrate that motions and interacting objects are highly correlated and that related object latent properties can be inferred from 3D skeleton sequences alone, leading to new synthesis  ...  Acknowledgements We sincerely thank the reviewers for their valuable comments. This work was supported in part by Shenzhen  ... 
doi:10.1007/s41095-021-0218-8 fatcat:3ign7zuhajgsjjavi3kjk6tqoq

Three-Dimensional Diffusion Model in Sports Dance Video Human Skeleton Detection and Extraction

Zhi Li, Miaochao Chen
2021 Advances in Mathematical Physics  
graph matching, thus proposing a matching algorithm for discrete skeleton points and optimizing it for the skeleton dislocation and algorithm problems of human occlusion.  ...  The research in this paper mainly includes as follows: for the principle of action recognition based on the 3D diffusion model convolutional neural network, the whole detection process is carried out from  ...  Human Skeleton Detection Based on THE 3D Diffusion Model Algorithm.  ... 
doi:10.1155/2021/3772358 fatcat:chfv5pithngubfcaqsbfwj2msm

GlocalNet: Class-aware Long-term Human Motion Synthesis [article]

Neeraj Battan, Yudhik Agrawal, Veeravalli Saisooryarao, Aman Goel, Avinash Sharma
2020 arXiv   pre-print
Synthesis of long-term human motion skeleton sequences is essential to aid human-centric video generation with potential applications in Augmented Reality, 3D character animations, pedestrian trajectory  ...  prediction, etc.  ...  In [32] , the authors proposed a method to generate human motion using a graph convolution network. RNN based approaches have performed well for action recognition, as shown in [20] .  ... 
arXiv:2012.10744v1 fatcat:icu7hy4tbjg23bofz7gk6mlwfq

Mapping the 3D Connectivity of the Rat Inner Retinal Vascular Network Using OCT Angiography

Conor Leahy, Harsha Radhakrishnan, Geoffrey Weiner, Jeffrey L. Goldberg, Vivek J. Srinivasan
2015 Investigative Ophthalmology and Visual Science  
The purpose of this study is to demonstrate three-dimensional (3D) graphing based on optical coherence tomography (OCT) angiography for characterization of the inner retinal vascular architecture and determination  ...  and human subjects.  ...  and human subjects.  ... 
doi:10.1167/iovs.15-17210 pmid:26325417 pmcid:PMC4559217 fatcat:jpe27ya7vbc47nsmydkdj42yku

Making the Invisible Visible: Action Recognition Through Walls and Occlusions [article]

Tianhong Li, Lijie Fan, Mingmin Zhao, Yingcheng Liu, Dina Katabi
2019 arXiv   pre-print
Our model takes radio frequency (RF) signals as input, generates 3D human skeletons as an intermediate representation, and recognizes actions and interactions of multiple people over time.  ...  In this paper, we introduce a neural network model that can detect human actions through walls and occlusions, and in poor lighting conditions.  ...  Also, some papers represented skeletons as graphs and utilized graph neural network (GNN) for action recognition [38, 13] .  ... 
arXiv:1909.09300v1 fatcat:zt4jrsra6navjojvmch5s4stei

MSST-RT: Multi-Stream Spatial-Temporal Relative Transformer for Skeleton-Based Action Recognition

Yan Sun, Yixin Shen, Liyan Ma
2021 Sensors  
Skeleton-based human action recognition has made great progress, especially with the development of a graph convolution network (GCN).  ...  Extensive experiments evaluate the proposed methods on three benchmarks for skeleton-based action recognition: NTU RGB+D, NTU RGB+D 120 and UAV-Human.  ...  and graph convolution-network-based methods.  ... 
doi:10.3390/s21165339 pmid:34450781 pmcid:PMC8401804 fatcat:tfcwfs6qsjhr5dj5nrorh7xro4

HOPE-Net: A Graph-based Model for Hand-Object Pose Estimation [article]

Bardia Doosti, Shujon Naha, Majid Mirbagheri, David Crandall
2020 arXiv   pre-print
The proposed 2D to 3D graph convolution-based model could be applied to other 3D landmark detection problems, where it is possible to first predict the 2D keypoints and then transform them to 3D.  ...  Our experiments show that through end-to-end training of the full network, we achieve better accuracy for both the 2D and 3D coordinate estimation problems.  ...  Acknowledgment The work in this paper was supported in part by the National Science Foundation (CAREER IIS-1253549), and by the IU Office of the Vice Provost for Research, the College of Arts and Sciences  ... 
arXiv:2004.00060v1 fatcat:63qxmkoyrbcrpdn72c4hfwxtmu

Adversarial Self-Supervised Learning for Semi-Supervised 3D Action Recognition [article]

Chenyang Si, Xuecheng Nie, Wei Wang, Liang Wang, Tieniu Tan, Jiashi Feng
2020 arXiv   pre-print
However, few effective self-supervised approaches exist for 3D action recognition, and directly applying SSL for semi-supervised learning suffers from misalignment of representations learned from SSL and  ...  Its major challenge lies in how to effectively learn motion representations from unlabeled data.  ...  For graph-structured data, graph-based approaches [31, 19, 32] are popularly adopted for skeleton-based action recognition, e.g., ST-GCN [44] and AGC-LSTM [30] .  ... 
arXiv:2007.05934v1 fatcat:5wfusiu7zzcsxdclrbonzkpqxm

AdaSGN: Adapting Joint Number and Model Size for Efficient Skeleton-Based Action Recognition [article]

Lei Shi, Yifan Zhang, Jian Cheng, Hanqing Lu
2021 arXiv   pre-print
Existing methods for skeleton-based action recognition mainly focus on improving the recognition accuracy, whereas the efficiency of the model is rarely considered.  ...  However, in addition to the model size, the amount of the data involved in the calculation is also an important factor for the running speed, especially for the skeleton data where most of the joints are  ...  Graph convolutional network (GCN) has been widely used for skeleton-based action recognition [38, 26] .  ... 
arXiv:2103.11770v1 fatcat:wfjdq2rlhfbvffeomsccbqpimy

Graph Matching for Marker Labelling and Missing Marker Reconstruction with Bone Constraint by LSTM in Optical Motion Capture

Jianfang Li, Degui Xiao, Keqin Li, Jiazhi Li
2021 IEEE Access  
First, a novel graph matching method is employed to determine the connection relationship of the scattered motion data for a single frame.  ...  The scattered reconstructed motion data must constitute a human configuration by labelling process according to the predefined template, and the missing markers have to be reconstructed to produce a stable  ...  ., for providing Cortex, Multiple Cameras Systems (MCSs), and 3D motion capture dataset.  ... 
doi:10.1109/access.2021.3060385 fatcat:6m4ooxmenfcprihfbg5p6624wa

SMART: Skeletal Motion Action Recognition aTtack [article]

He Wang, Feixiang He, Zhexi Peng, Yongliang Yang, Tianjia Shao, Kun Zhou, David Hogg
2020 arXiv   pre-print
In this paper, we propose a method, SMART, to attack action recognizers which rely on 3D skeletal motions.  ...  Finally, SMART shows that adversarial attack on 3D skeletal motion, one type of time-series data, is significantly different from traditional adversarial attack problems.  ...  Recent advances in 3D sensing and pose estimation motivate the use of clean skeleton data to robustly classify human actions, overcoming the biases from raw RGB video due to body occlusion, scattered background  ... 
arXiv:1911.07107v3 fatcat:eurao6iimzgzpj7kvnqprc5qyu

Semantics-guided Skeletonization of Sweet Cherry Trees for Robotic Pruning [article]

Alexander You, Cindy Grimm, Abhisesh Silwal, Joseph R. Davidson
2021 arXiv   pre-print
We test our skeletonization algorithm on point clouds from 29 upright fruiting offshoot (UFO) trees and demonstrate a median accuracy of 70% with respect to a human-evaluated gold standard.  ...  One useful structure for modeling a tree is a skeleton: a 1D, lightweight representation of the geometry and the topology of a tree.  ...  It returns two prediction vectors: One for whether or not the corresponding edge is valid, and one predicting whether the edge is a trunk, support, leader, side branch, or something else. 2: Population-based  ... 
arXiv:2103.02833v1 fatcat:wh4hmp4pnfg2dmsdzkeqbgemka

Facial Landmark-Based Emotion Recognition via Directed Graph Neural Network

Quang Tran Ngoc, Seunghyun Lee, Byung Cheol Song
2020 Electronics  
In this paper, we propose a graph convolution neural network that utilizes landmark features for FER, which we called a directed graph neural network (DGNN).  ...  Also, a fusion network using image information as well as landmarks, is presented and investigated for the CK+ (98.47% performance) and AFEW (50.65% performance) datasets.  ...  Each skeleton point corresponds to a joint on a human body, and skeletons can be represented by a graph structure.  ... 
doi:10.3390/electronics9050764 fatcat:bp73dqwbdrddxmnjofik6dnoxu

Enhancing Egocentric 3D Pose Estimation with Third Person Views [article]

Ameya Dhamanaskar, Mariella Dimiccoli, Enric Corona, Albert Pumarola, Francesc Moreno-Noguer
2022 arXiv   pre-print
Our dataset and code will be available for research purposes.  ...  We explicitly consider spatial- and motion-domain features, combined using a semi-Siamese architecture trained in a self-supervised fashion.  ...  Motion-Graph [4] is currently the state-of-the art method for predicting 3D body pose from real egocentric videos without a second interacting person.  ... 
arXiv:2201.02017v2 fatcat:soz7m72sb5altjx7of7shelz2u
« Previous Showing results 1 — 15 out of 900 results