2,929 Hits in 5.8 sec

Gradient local auto-correlation features for depth human action recognition

Mohammad Farhad Bulbul, Hazrat Ali
2021 SN Applied Sciences  
In the approach, the enhanced motion and static history images are computed and a set of 2D auto-correlation gradient feature vectors is obtained from them to describe an action.  ...  Kernel-based Extreme Learning Machine is used with the extracted features to distinguish the diverse action types promisingly.  ...  In addition, we can obtain body shape and structure characteristics and the human skeleton information from depth images.  ... 
doi:10.1007/s42452-021-04528-1 fatcat:zpumt5f25fdydlowttzcqjrm4e

Stacked sparse autoencoder and history of binary motion image for human activity recognition

Mariem Gnouma, Ammar Ladjailia, Ridha Ejbali, Mourad Zaied
2018 Multimedia tools and applications  
In this paper, a supervised way followed by an unsupervised learning using the principle of the auto-encoder is proposed to address the problem.  ...  Tools Appl (SSAE), an instance of a deep learning strategy, is presented for efficient human activities detection and the Softmax (SMC) for the classification.  ...  Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.  ... 
doi:10.1007/s11042-018-6273-1 fatcat:jiltrfaj3nbubg2q2qm6ivjgtu

Structured Prediction of 3D Human Pose with Deep Neural Networks [article]

Bugra Tekin, Isinsu Katircioglu, Mathieu Salzmann, Vincent Lepetit, Pascal Fua
2016 arXiv   pre-print
In this paper, we introduce a Deep Learning regression architecture for structured prediction of 3D human pose from monocular images that relies on an overcomplete auto-encoder to learn a high-dimensional  ...  They either train a Convolutional Neural Network to directly regress from image to 3D pose, which ignores the dependencies between human joints, or model these dependencies via a max-margin structured  ...  Using auto-encoders for unsupervised feature learning has proven effective in several recognition tasks [16, 18, 35] .  ... 
arXiv:1605.05180v1 fatcat:mhpcvbkmsvefrfqn2q44gqdx3m

Deep Learning Methods for Cardiovascular Image

Yankun Cao, Zhi Liu, Pengfei Zhang, Yushuo Zheng, Yongsheng Song, Lizhen Cui
2019 Journal of Artificial Intelligence and Systems  
Cardiovascular disease is one of the most important diseases that endanger human health at present. It is very meaningful to diagnose and treat cardiovascular disease by means of in-depth learning.  ...  learning, and then summarizes the application of deep learning in heart image segmentation, classification and other aspects combined with existing technologies.  ...  Here, the authors need to declare whether or not the submitted work was carried out in the presence of any personal, professional or financial relationships that could potentially be construed as a conflict  ... 
doi:10.33969/ais.2019.11006 fatcat:egx5tibvm5dhriemz4dvx5ktsm

VR content creation and exploration with deep learning: A survey

Miao Wang, Xu-Quan Lyu, Yi-Jun Li, Fang-Lue Zhang
2020 Computational Visual Media  
This article surveys recent research that uses such deep learning methods for VR content creation and exploration.  ...  Intelligence of VR methods and applications has been significantly boosted by the recent developments in deep learning techniques.  ...  Here, we review existing deep learning-based work for several major categories of reconstruction methods, namely for general scenes, human faces, and human bodies.  ... 
doi:10.1007/s41095-020-0162-z fatcat:lgogzx26bvhn5f7uyefjkz7zny

2021 Index IEEE Transactions on Multimedia Vol. 23

2021 IEEE transactions on multimedia  
The Author Index contains the primary entry for each item, listed under the first author's name.  ...  -that appeared in this periodical during 2021, and items from previous years that were commented upon or corrected in 2021.  ...  ., +, TMM 2021 4388-4399 Image denoising Adversarial 3D Convolutional Auto-Encoder for Abnormal Event Detection in Videos.  ... 
doi:10.1109/tmm.2022.3141947 fatcat:lil2nf3vd5ehbfgtslulu7y3lq

Autonomous Driving with Deep Learning: A Survey of State-of-Art Technologies [article]

Yu Huang, Yue Chen
2020 arXiv   pre-print
Due to the limited space, we focus the analysis on several key areas, i.e. 2D and 3D object detection in perception, depth estimation from cameras, multiple sensor fusion on the data, feature and task  ...  Almost at the same time, deep learning has made breakthrough by several pioneers, three of them (also called fathers of deep learning), Hinton, Bengio and LeCun, won ACM Turin Award in 2019.  ...  a layer called sparse non-homogeneous pooling layer to transform features between bird's eye view and front view based on point cloud, where the encoder backbone for RGB image and LiDAR data is MSCNN  ... 
arXiv:2006.06091v3 fatcat:nhdgivmtrzcarp463xzqvnxlwq

Learning Gaze Transitions from Depth to Improve Video Saliency Estimation [article]

G. Leifman, D. Rudoy, T. Swedish, E. Bayro-Corrochano, R. Raskar
2016 arXiv   pre-print
In this paper we introduce a novel Depth-Aware Video Saliency approach to predict human focus of attention when viewing RGBD videos on regular 2D screens.  ...  Our experiments indicate that integrating depth into video saliency calculation is beneficial.  ...  [26] present a depth prior for saliency learned from human gaze information.  ... 
arXiv:1603.03669v1 fatcat:nezm5psydvf7zbf5ouph7kodee

Action recognition in videos

Christian Wolf, Atilla Baskurt
2012 2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)  
We briefly describe two recent contributions: joint level feature and sequence learning, as well as space-time graph matching.  ...  Applications such as video surveillance, robotics, source selection, and video indexing often require the recognition of actions based on the motion of different actors in a video.  ...  In this context, and as an extension of work in object recognition [9] , we proposed a translation variant sparse convolutional auto-encoder, which automatically learns a sparse feature extractor [1]  ... 
doi:10.1109/ipta.2012.6469480 dblp:conf/ipta/WolfB12 fatcat:627gostylvhtji5tvnsehvj4pq

Person identification from streaming surveillance video using mid-level features from joint action-pose distribution

Binu M. Nair, Vijayan K. Asari, Robert P. Loce, Eli Saber
2015 Video Surveillance and Transportation Imaging Applications 2015  
We propose a real time person identification algorithm for surveillance based scenarios from low-resolution streaming video, based on mid-level features extracted from the joint distribution of various  ...  The proposed algorithm uses the combination of an auto-encoder based action association framework which produces per-frame probability estimates of the action being performed, and a pose recognition framework  ...  Research in human action recognition can be grouped into different categories such as image models, sparse features and grammars/templates. 1 Some state of the art algorithms under the banner of image  ... 
doi:10.1117/12.2083423 fatcat:q3r7rqa5snffpanwvnaqcmbxzm

RGB-D-based Human Motion Recognition with Deep Learning: A Survey [article]

Pichao Wang and Wanqing Li and Philip Ogunbona and Jun Wan and Sergio Escalera
2018 arXiv   pre-print
In particular, convolutional neural networks (CNN) have achieved great success for image-based tasks, and recurrent neural networks (RNN) are renowned for sequence-based problems.  ...  The reviewed methods are broadly categorized into four groups, depending on the modality adopted for recognition: RGB-based, depth-based, skeleton-based and RGB+D-based.  ...  Ijjina et al. [57] adopted stacked auto encoder to learn the underlying features of input skeleton data.  ... 
arXiv:1711.08362v2 fatcat:cugugpqeffcshnwwto4z2aw4ti

Human Pose Estimation from Monocular Images: A Comprehensive Survey

Wenjuan Gong, Xuena Zhang, Jordi Gonzàlez, Andrews Sobral, Thierry Bouwmans, Changhe Tu, El-hadi Zahzah
2016 Sensors  
The human pose detection problem has seen the most success when utilizing depth images in conjunction with color images: real-time estimation of 3D body joints and pixelwise body part labelling have been  ...  Furthermore, recent advancements based on deep learning have brought novel algorithms for this problem.  ...  In part-based methods, body part candidates are first detected from image evidence, and then detected body parts are assembled to fit image observations and a body plan [206] .  ... 
doi:10.3390/s16121966 pmid:27898003 pmcid:PMC5190962 fatcat:jigvz4ovpbh63eovto3etoefx4

Medical Image Fusion Method by Deep Learning

Yi Li, Junli Zhao, Zhihan Lv, Jinhua Li
2021 International Journal of Cognitive Computing in Engineering  
Deep learning technology has been extensively explored in pattern recognition and image processing areas.  ...  It cannot be only made up for the deficiencies of MRI, CT and SPECT image fusion, but also can be implemented in different types of multi-modal medical image fusion problems in batch processing mode, and  ...  Acknowledgments The authors first sincerely thank the editors and anonymous reviewers for their constructive comments and suggestions, which are of great value to us.  ... 
doi:10.1016/j.ijcce.2020.12.004 fatcat:aeuelnzwlzdlld66lz334ivhwm

Rendering Portraitures from Monocular Camera and Beyond [chapter]

Xiangyu Xu, Deqing Sun, Sifei Liu, Wenqi Ren, Yu-Jin Zhang, Ming-Hsuan Yang, Jian Sun
2018 Lecture Notes in Computer Science  
In addition, we train a spatially-variant Recursive Neural Network to learn and accelerate this rendering process.  ...  In this work, we introduce an automatic system that achieves portrait DoF rendering for monocular cameras.  ...  Depth Estimation with Single Image. Deep learning based models have been used to learn depth from a single image, both in supervised and unsupervised ways. For supervised depth learning, Eigen et al.  ... 
doi:10.1007/978-3-030-01240-3_3 fatcat:wrjphnnkozdjbjdxoxqkpdcdni

Deep learning for detecting robotic grasps

Ian Lenz, Honglak Lee, Ashutosh Saxena
2015 The international journal of robotics research  
In this work, we apply a deep learning approach to solve this problem, which avoids time-consuming hand-design of features. This presents two main challenges.  ...  Second, we need to handle multimodal inputs effectively, for which we present a method that applies structured regularization on the weights based on multimodal group regularization.  ...  ACKNOWLEDGEMENTS We would like to thank Yun Jiang and Marcus Lim for useful discussions and help with baseline experiments.  ... 
doi:10.1177/0278364914549607 fatcat:vgo22gpforgb5mtyi4ogl27eny
« Previous Showing results 1 — 15 out of 2,929 results