Filters








1,441 Hits in 6.2 sec

Body Joints and Trajectory Guided 3D Deep Convolutional Descriptors for Human Activity Identification

2019 VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE  
Therefore, the key goal of this work is to improve the recognition accuracy using an optical flow integrated with a two-stream bilinear model, namely Joints and Trajectory-pooled 3D-Deep convolutional  ...  Among various HAI techniques, Joints-pooled 3D-Deep convolutional Descriptors (JDD) have achieved effective performance by learning the body joint and capturing the spatiotemporal characteristics concurrently  ...  [13] proposed an action recognition model with the help of Trajectory-pooled Deep-convolutional Descriptor (TDD).  ... 
doi:10.35940/ijitee.k1985.1081219 fatcat:ilirq3zbfjcmrab7b6oxpoprse

Body Joint guided 3D Deep Convolutional Descriptors for Action Recognition [article]

Congqi Cao, Yifan Zhang, Chunjie Zhang, Hanqing Lu
2017 arXiv   pre-print
to video-based action recognition.  ...  The helpfulness of the body joint guided feature pooling with inaccurate skeleton estimation is systematically evaluated.  ...  JOINTS-POOLED 3D DEEP CONVOLUTIONAL DESCRIPTORS In this section, we give an introduction to the proposed joints-pooled 3D deep convolutional descriptors (JDD).  ... 
arXiv:1704.07160v2 fatcat:k6agf72it5addp5vdslleecyy4

A Survey on Deep Learning Based Approaches for Action and Gesture Recognition in Image Sequences

Maryam Asadi-Aghbolaghi, Albert Clapes, Marco Bellantonio, Hugo Jair Escalante, Victor Ponce-Lopez, Xavier Baro, Isabelle Guyon, Shohreh Kasaei, Sergio Escalera
2017 2017 12th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2017)  
In this paper, we present a survey on current deep learning methodologies for action and gesture recognition in image sequences.  ...  The interest in action and gesture recognition has grown considerably in the last years.  ...  [12] , takes advantage of human body spatial constraints, by aggregating convolutional activations of a 3D CNN into descriptors based on joint positions.  ... 
doi:10.1109/fg.2017.150 dblp:conf/fgr/Asadi-Aghbolaghi17 fatcat:wzkf5sfc5ncsfjicmkfuw4owxq

Deep Learning for Action and Gesture Recognition in Image Sequences: A Survey [chapter]

Maryam Asadi-Aghbolaghi, Albert Clapés, Marco Bellantonio, Hugo Jair Escalante, Víctor Ponce-López, Xavier Baró, Isabelle Guyon, Shohreh Kasaei, Sergio Escalera
2017 Gesture Recognition  
A survey on deep learning based approaches for action and gesture recognition in image sequences.  ...  This chapter is a survey of current deep learning based methodologies for action and gesture recognition in sequences of images.  ...  Wang et al. (2015b) introduce a video representation called Trajectory-pooled Deep-convolutional Descriptor (TDD), which consists on extending the state-of-the-art descriptors along the trajectories with  ... 
doi:10.1007/978-3-319-57021-1_19 fatcat:d2m5oyomsjhkbfpunhefho6ayq

Going Deeper into Action Recognition: A Survey [article]

Samitha Herath, Mehrtash Harandi, Fatih Porikli
2017 arXiv   pre-print
To this end, we start our discussion with the pioneering methods that use handcrafted representations, and then, navigate into the realm of deep learning based approaches.  ...  Understanding human actions in visual data is tied to advances in complementary research areas including object recognition, human dynamics, domain adaptation and semantic segmentation.  ...  Considering deep architectures for action recognition, the keywords to remember would be 3D convolutions, temporal pooling, optical flow frames, and LSTMs.  ... 
arXiv:1605.04988v2 fatcat:7727tjctgfffzlnig5rvicxjgq

Handcrafted vs. learned representations for human action recognition

Xiantong Zhen, Ling Shao, Stephen J. Maybank, Rama Chellappa
2016 Image and Vision Computing  
The article "3D-based Deep Convolutional Neural Network for action recognition with depth sequences" introduces a new deep learning based method for action recognition in depth sequences.  ...  In this work, a 3D-based Deep Convolutional Neural Network (3D2CNN) to directly learn spatiotemporal features from raw depth sequences.  ... 
doi:10.1016/j.imavis.2016.10.002 fatcat:j4c2txj3g5glra67qvzab5mmke

Two-Stream 3D Convolutional Neural Network for Skeleton-Based Action Recognition [article]

Hong Liu and Juanhui Tu and Mengyuan Liu
2017 arXiv   pre-print
It remains a challenge to efficiently extract spatialtemporal information from skeleton sequences for 3D human action recognition.  ...  To our best knowledge, this is the first application of 3D CNN in skeleton-based action recognition. Our method consists of three stages.  ...  [21] provided an more effective and robust joints-pooled 3D deep convolutional H descriptor (JDD), generating promising results on real-world datasets.  ... 
arXiv:1705.08106v2 fatcat:lebbvcc4qjbhtfqwnafcz5nyde

Human Action Recognition Based on Temporal Pose CNN and Multi-dimensional Fusion [chapter]

Yi Huang, Shang-Hong Lai, Shao-Heng Tai
2019 Landolt-Börnstein - Group III Condensed Matter  
To take advantage of recent advances in human pose estimation from images, we develop a deep neural network model for action recognition from videos by computing temporal human pose features with a 3D  ...  We show that the proposed action recognition system provides superior accuracy compared to the previous methods through experiments on Sub-JHMDB and PennAction datasets.  ...  [2] proposed to pool 3D deep CNN activations of different segments of a video using joint positions of frames in the video.  ... 
doi:10.1007/978-3-030-11012-3_33 fatcat:nz3q2f7hrjeqjoszc6iam2a4zu

Self-Intelligence with Human Activities Recognition based on Convolutional Neural Networks

2020 International Journal of Engineering and Advanced Technology  
In the presented paper, we propose a strategy related to activity recognition of human from profundity maps as well as sequences stance information using convolutional neural systems.  ...  Two information descriptors will be utilized for activity portrayal.  ...  A.Benchmark Datasets MSR 3D The most widely used dataset used for the recognition of action is MSRAction3D.  ... 
doi:10.35940/ijeat.d6489.049420 fatcat:sbqqxt5qjfgfvdczlq2vo55lcu

Human Behavior Analysis: A Survey on Action Recognition

Bruno Degardin, Hugo Proença
2021 Applied Sciences  
The visual recognition and understanding of human actions remain an active research domain of computer vision, being the scope of various research works over the last two decades.  ...  Previous surveys mainly focus on the evolution of this field, from handcrafted features to deep learning architectures.  ...  [11] a spatio-temporal feature learning by using deep convolutional 3dimensional networks (3D ConvNets).  ... 
doi:10.3390/app11188324 fatcat:zenvfhlaubht7ar3qrpil4lgdm

Convolutional Long Short-Term Memory Networks for Recognizing First Person Interactions [article]

Swathikiran Sudhakaran, Oswald Lanz
2017 arXiv   pre-print
In this paper, we present a novel deep learning based approach for addressing the problem of interaction recognition from a first person perspective.  ...  In particular, on UTKinect-FirstPerson it competes with methods that use depth image and skeletal joints information along with RGB images, while it surpasses all previous methods that use only RGB images  ...  Acknowledgments We gratefully acknowledge the support of NVIDIA Corporation with the donation of GPU used for this research.  ... 
arXiv:1709.06495v1 fatcat:ikslmyzh2jednpwprc3ra6x5wi

Convolutional Long Short-Term Memory Networks for Recognizing First Person Interactions

Swathikiran Sudhakaran, Oswald Lanz
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
In this paper, we present a novel deep learning based approach for addressing the problem of interaction recognition from a first person perspective.  ...  In particular, on UTKinect-FirstPerson it competes with methods that use depth image and skeletal joints information along with RGB images, while it surpasses all previous methods that use only RGB images  ...  Acknowledgments We gratefully acknowledge the support of NVIDIA Corporation with the donation of GPU used for this research.  ... 
doi:10.1109/iccvw.2017.276 dblp:conf/iccvw/SudhakaranL17 fatcat:dmkd6noojjfvvekouzehqtuecy

Skeleton based Activity Recognition by Fusing Part-wise Spatio-temporal and Attention Driven Residues [article]

Chhavi Dhiman, Dinesh Kumar Vishwakarma, Paras Aggarwal
2019 arXiv   pre-print
In this paper, we present a novel skeleton-based part-wise Spatiotemporal CNN RIAC Network-based 3D human action recognition framework to visualise the action dynamics in part wise manner and utilise each  ...  To extract and learn salient features for action recognition, attention driven residues are used which enhance the performance of residual components for effective 3D skeleton-based Spatio-temporal action  ...  The proposed work defines a novel action descriptor by combining attention based residues with Spatial-Temporal based Convolution Features (STCF).  ... 
arXiv:1912.00576v1 fatcat:4pg77axdxbd43p6sa6lt6fmnoe

Literature Review of Action Recognition in the Wild [article]

Asket Kaur, Navya Rao, Tanya Joon
2019 arXiv   pre-print
Action Recognition problem in the untrimmed videos is a challenging task and most of the papers have tackled this problem using hand-crafted features with shallow learning techniques and sophisticated  ...  The literature review presented below on Action Recognition in the wild is the in-depth study of Research Papers.  ...  Overview The authors of this paper investigated architectures that are trained on deep Convolutional Networks for action recognition in video.  ... 
arXiv:1911.12249v1 fatcat:46qu4wtyqvhuxcomoymdd5owcm

First Person Action Recognition Using Deep Learned Descriptors

Suriya Singh, Chetan Arora, C. V. Jawahar
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We propose convolutional neural networks (CNNs) for end to end learning and classification of wearer's actions.  ...  We focus on the problem of wearer's action recognition in first person a.k.a. egocentric videos.  ...  [44] extended 2D convolution to allow 3D convolution and 3D pooling, which has been shown to capture the temporal structure of an action.  ... 
doi:10.1109/cvpr.2016.287 dblp:conf/cvpr/SinghAJ16 fatcat:xy7qjd3cvbgydmchfwv2upz3ju
« Previous Showing results 1 — 15 out of 1,441 results