Filters








4,506 Hits in 5.7 sec

Unsupervised Feature Learning for RGB-D Image Classification [chapter]

I-Hong Jhuo, Shenghua Gao, Liansheng Zhuang, D. T. Lee, Yi Ma
2015 Lecture Notes in Computer Science  
Motivated by the success of Deep Neural Networks in computer vision, we propose a deep Regularized Reconstruction Independent Component Analysis network (R 2 ICA) for RGB-D image classification.  ...  Implementing commonly used local contrast normalization and spatial pooling, we gradually enhance our network to be resilient to local variance resulting in a robust image representation for RGB-D image  ...  neural network for RGB-D image classification. 1 The resultant deep neural network boosts both the accuracy and efficiency for RGB-D image classification; (2) we propose the R 2 ICA algorithm and implement  ... 
doi:10.1007/978-3-319-16865-4_18 fatcat:zukahvjhenfsnfson4y3bpbrem

2020 Index IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 42

2021 IEEE Transactions on Pattern Analysis and Machine Intelligence  
., +, TPAMI May 2020 1228-1242 Robust RGB-D Face Recognition Using Attribute-Aware Loss. Jiang, L., +, Squeeze-and-Excitation Networks.  ...  ., +, 2508- 2522 Unsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery.  ...  Object recognition Adversarial Action Prediction Networks. Kong, Y., +, TPAMI March 2020 539-553 Capturing the Geometry of Object Categories from Video Supervision.  ... 
doi:10.1109/tpami.2020.3036557 fatcat:3j6s2l53x5eqxnlsptsgbjeebe

Object recognition with hierarchical kernel descriptors

Liefeng Bo, Kevin Lai, Xiaofeng Ren, Dieter Fox
2011 CVPR 2011  
We evaluate hierarchical kernel descriptors both on the CIFAR10 dataset and the new RGB-D Object Dataset consisting of segmented RGB and depth images of 300 everyday objects.  ...  Kernel descriptors [1] provide a unified way to generate rich visual feature sets by turning pixel attributes into patch-level features, and yield impressive results on many object recognition tasks.  ...  RGB-D Object Dataset We evaluated hierarchical kernel descriptors on the RGB-D Object Dataset.  ... 
doi:10.1109/cvpr.2011.5995719 dblp:conf/cvpr/BoLRF11 fatcat:jv4iqtjl4vhefoxeozyhlwfhgy

RGB-D Data-Based Action Recognition: A Review

Muhammad Bilal Shaikh, Douglas Chai
2021 Sensors  
In this paper, we focus solely on data fusion and recognition techniques in the context of vision with an RGB-D perspective.  ...  Naturally, each action-data modality—such as RGB, depth, skeleton, and infrared (IR)—has distinct characteristics; therefore, it is important to exploit the value of each modality for better action recognition  ...  Acknowledgments: The authors would like to thank the anonymous reviewers for their careful reading and valuable remarks, which have greatly helped extend the scope of this paper.  ... 
doi:10.3390/s21124246 fatcat:7dvocdy63rckne5yunhfsnr4p4

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.  ...  Specifically, deep learning methods based on the CNN and RNN architectures have been adopted for motion recognition using RGB-D data.  ...  Figure 1 : 1 Categorisation of the methods for RGB-D-based motion recognition using deep learning. congd.html) is a large RGB-D dataset for continuous gesture recognition.  ... 
arXiv:1711.08362v2 fatcat:cugugpqeffcshnwwto4z2aw4ti

Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks [article]

Hongsong Wang, Liang Wang
2017 arXiv   pre-print
Recent methods that use Recurrent Neural Networks (RNN) to handle raw skeletons only focus on the contextual dependency in the temporal domain and neglect the spatial configurations of articulated skeletons  ...  In this paper, we propose a novel two-stream RNN architecture to model both temporal dynamics and spatial configurations for skeleton based action recognition.  ...  Inspired by the great success of deep learning for RGB based action recognition [39, 26, 21] , there is a growing trend of using deep neural networks for skeleton based action recognition.  ... 
arXiv:1704.02581v2 fatcat:lhpg3yevkbctdh44ncbzry2dxm

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.  ...  We introduce a taxonomy that summarizes important aspects of deep learning for approaching both tasks.  ...  A. 3D Convolutional Neural Networks Several 3D CNNs have been proposed for gesture recognition, most notably [64, 41, 63] . [41] proposes a 3D CNN for sign language recognition.  ... 
doi:10.1109/fg.2017.150 dblp:conf/fgr/Asadi-Aghbolaghi17 fatcat:wzkf5sfc5ncsfjicmkfuw4owxq

Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks

Hongsong Wang, Liang Wang
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Recent methods that use Recurrent Neural Networks (RNN) to handle raw skeletons only focus on the contextual dependency in the temporal domain and neglect the spatial configurations of articulated skeletons  ...  In this paper, we propose a novel two-stream RNN architecture to model both temporal dynamics and spatial configurations for skeleton based action recognition.  ...  Action recognition with deep networks Deep neural networks have made great progress in the area of action recognition. 3D Convolutional Neural Networks (CNN) is proposed and different architectures are  ... 
doi:10.1109/cvpr.2017.387 dblp:conf/cvpr/WangW17 fatcat:izmydc7rdvh3pcsbu4k4wxb32a

Recent advances in video-based human action recognition using deep learning: A review

Di Wu, Nabin Sharma, Michael Blumenstein
2017 2017 International Joint Conference on Neural Networks (IJCNN)  
This paper presents a review of various state-of-theart deep learning-based techniques proposed for human action recognition on the three types of datasets.  ...  Three different types of datasets namely, single viewpoint, multiple viewpoint and RGB-depth videos, are used for research.  ...  Ali and Wang [58] built a Deep Brief Network (DBN) which is another variant of deep neural networks.  ... 
doi:10.1109/ijcnn.2017.7966210 dblp:conf/ijcnn/WuSB17 fatcat:f35o5nkxozfsrew2sgtaybofla

Convolutional-Recursive Deep Learning for 3D Object Classification

Richard Socher, Brody Huval, Bharath Putta Bath, Christopher D. Manning, Andrew Y. Ng
2012 Neural Information Processing Systems  
We introduce a model based on a combination of convolutional and recursive neural networks (CNN and RNN) for learning features and classifying RGB-D images.  ...  Our model obtains state of the art performance on a standard RGB-D object dataset while being more accurate and faster during training and testing than comparable architectures such as two-layer CNNs.  ...  FA8750-09-C-0181, and the DARPA Deep Learning program under contract number FA8650-10-C-7020.  ... 
dblp:conf/nips/SocherHBMN12 fatcat:52b6veja5zbfffbujjkutfuuxu

Compact Deep Color Features for Remote Sensing Scene Classification

Rao Muhammad Anwer, Fahad Shahbaz Khan, Jorma Laaksonen
2021 Neural Processing Letters  
We show that combining several deep color models significantly improves the recognition performance compared to using the RGB network alone.  ...  Most recent aerial scene classification approaches are based on Convolutional Neural Networks (CNNs).  ...  Top row: RGB space-based deep convolutional neural network (left) and Opponent color space-based deep convolutional neural network (right).  ... 
doi:10.1007/s11063-021-10463-4 fatcat:getvq2myhvayhbgzdzpknzenkq

RGB-D-Based Object Recognition Using Multimodal Convolutional Neural Networks: A Survey

Mingliang Gao, Jun Jiang, Guofeng Zou, Vijay John, Zheng Liu
2019 IEEE Access  
INDEX TERMS Convolutional neural network, multimodal fusion, object recognition, RGB-D, survey. 43110 2169-3536 2019 IEEE. Translations and content mining are permitted for academic research only.  ...  RGB-D-based object recognition has evolved from early methods that using hand-crafted representations to the current state-of-the-art deep learning-based methods.  ...  [205] proposed analog backpropagation learning circuits for various memristive learning architectures, such as deep neural network, binary neural network, multiple neural network, hierarchical temporal  ... 
doi:10.1109/access.2019.2907071 fatcat:shamfnufhfavjlcnrcldpgqtgq

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.  ...  For gesture recognition from RGB-D data Duan et al. (2016) use two general deep-based network; i.e., convolutional two stream consensus voting network (2SCVN) for modeling the RGB and optical flow and  ... 
doi:10.1007/978-3-319-57021-1_19 fatcat:d2m5oyomsjhkbfpunhefho6ayq

RGB-D Object Recognition Using Deep Convolutional Neural Networks

Saman Zia, Buket Yuksel, Deniz Yuret, Yucel Yemez
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
We address the problem of object recognition from RGB-D images using deep convolutional neural networks (CNNs).  ...  To this end, we propose a hybrid 2D/3D convolutional neural network that can be initialized with pretrained 2D CNNs and can then be trained over a relatively small RGB-D dataset.  ...  Convolutional neural networks have been used in combination with other architectures to solve the RGB-D object recognition problem.  ... 
doi:10.1109/iccvw.2017.109 dblp:conf/iccvw/ZiaYYY17 fatcat:atpqmvnuo5d2bciwxjmy5kp7jm

Multi Modal RGB D Action Recognition with CNN LSTM Ensemble Deep Network

D. Srihari, P. V.
2020 International Journal of Advanced Computer Science and Applications  
The objective of this work is to perform multi modal human action recognition on an ensemble hybrid network of CNN and LSTM layers.  ...  The hybrid network is found to be receptive towards both spatial and temporal fields because of the hierarchical structure of CNNs and LSTMs.  ...  These multi-dimensional tensors are processed through deep convolutional neural networks (CNN) for learning spatial representations thereby completely ignoring the temporal structures [7] .  ... 
doi:10.14569/ijacsa.2020.0111284 fatcat:h63esrv6pfhljkzt7xdy6ygypa
« Previous Showing results 1 — 15 out of 4,506 results