64,505 Hits in 6.8 sec

Very Deep Convolutional Networks for Large-Scale Image Recognition [article]

Karen Simonyan, Andrew Zisserman
2015 arXiv   pre-print
In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting.  ...  Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art  ...  network depth on its accuracy in the large-scale image recognition setting.  ... 
arXiv:1409.1556v6 fatcat:o5faxhgt45bmzfyrjyuzmxnzmu

Super-Resolution using Deep Learning to Support Person Identification in Surveillance Video

Lamya Alkanhal, Deena Alotaibi, Nada Albrahim, Sara Alrayes, Ghaida Alshemali, Ouiem Bchir
2020 International Journal of Advanced Computer Science and Applications  
More specifically, we used the Very-Deep Super-Resolution (VDSR) neural network to enhance the image quality.  ...  Most importantly, it increased the face recognition rate by 45.7%.  ...  VERY DEEP SUPER RESOLUTION NETWORK STRUCTURE The Very Deep Super Resolution, VDSR, is a Convolutional Neural Network (CNN). It is designed to solve the SR problem for generic images [8] .  ... 
doi:10.14569/ijacsa.2020.0110749 fatcat:2tq5y3hujbharowhtj2h4cjsp4

MobiFace: A Lightweight Deep Learning Face Recognition on Mobile Devices [article]

Chi Nhan Duong, Kha Gia Quach, Ibsa Jalata, Ngan Le, Khoa Luu
2019 arXiv   pre-print
It is also eventually competitive against large-scale deep-networks face recognition while significant reducing computational time and memory consumption.  ...  In this paper, a novel deep neural network named MobiFace, a simple but effective approach, is proposed for productively deploying face recognition on mobile devices.  ...  Although the model is very small, its performance on both testing benchmarks is competitive against other large-scale deep face recognition network.  ... 
arXiv:1811.11080v2 fatcat:yoh3a5grcbavhl5l2hqm5dzfsq

Learning Face Representation from Scratch [article]

Dong Yi, Zhen Lei, Shengcai Liao, Stan Z. Li
2014 arXiv   pre-print
Pushing by big data and deep convolutional neural network (CNN), the performance of face recognition is becoming comparable to human.  ...  Using private large scale training datasets, several groups achieve very high performance on LFW, i.e., 97% to 99%.  ...  The Tesla K40 used for this research was donated by the NVIDIA Corporation.  ... 
arXiv:1411.7923v1 fatcat:pcezmc7hbbfsldd3wz4x7wr3ou

Deep Face Recognition

Omkar M. Parkhi, Andrea Vedaldi, Andrew Zisserman
2015 Procedings of the British Machine Vision Conference 2015  
Recent progress in this area has been due to two factors: (i) end to end learning for the task using convolutional neural networks (CNNs), and (ii) the availability of very large scale training datasets  ...  the trade-off between annotation purity and cost; second, we introduce a very deep convolutional neural network and a corresponding training procedure that achieve face recognition accuracy comparable  ...  Recent progress in this area has been due to two factors: (i) end to end learning for the task using convolutional neural networks (CNNs), and (ii) the availability of very large scale training datasets  ... 
doi:10.5244/c.29.41 dblp:conf/bmvc/ParkhiVZ15 fatcat:vhncyzl2dbd5fjmf6rama7oqia

LOGO-Net: Large-scale Deep Logo Detection and Brand Recognition with Deep Region-based Convolutional Networks [article]

Steven C.H. Hoi, Xiongwei Wu, Hantang Liu, Yue Wu, Huiqiong Wang, Hui Xue, Qiang Wu
2015 arXiv   pre-print
deep region-based convolutional networks techniques for object detection tasks.  ...  In this paper, we introduce "LOGO-Net", a large-scale logo image database for logo detection and brand recognition from real-world product images.  ...  Conclusions This paper presented "LOGO-Net" -a large-scale logo image database to facilitate large-scale deep logo detection and brand recognition from real-world product images.  ... 
arXiv:1511.02462v2 fatcat:km4x5wku4nhw3ptqcz76gpznge

Deep Learning Algorithms with Applications to Video Analytics for A Smart City: A Survey [article]

Li Wang, Dennis Sng
2015 arXiv   pre-print
Deep learning has recently achieved very promising results in a wide range of areas such as computer vision, speech recognition and natural language processing.  ...  In this paper, we review the deep learning algorithms applied to video analytics of smart city in terms of different research topics: object detection, object tracking, face recognition, image classification  ...  First, around 2000 region proposals are generated within the input image. For each proposal, a large convolutional neural network is used to extract features.  ... 
arXiv:1512.03131v1 fatcat:kavsqti6nvh6lnkz62tk7adtu4

Aging Face Recognition Using Deep Learning

Yogita Mahajan, Shanta Sondur
2018 International Journal of Engineering and Applied Sciences (IJEAS)  
This work focuses on the aging face recognition problems of entire face image based on a deep learning method, in particular, convolutional neural network.  ...  56  Abstract-Deep learning based approaches has gained very optimistic results in face recognition area.  ...  This paper focused on Aging Face Recognition problem by using Convolutional Neural Network, named as AFR-CNN. Deep learning using CNN has become very popular nowadays.  ... 
doi:10.31873/ijeas.5.8.17 fatcat:uhveqmymhbfdnkfgpbz37x77wu

Celebrity Face Recognition using Deep Learning

Nur Ateqah Binti Mat Kasim, Nur Hidayah Binti Abd Rahman, Zaidah Ibrahim, Nur Nabilah Abu Mangshor
2018 Indonesian Journal of Electrical Engineering and Computer Science  
One of the techniques under deep learning is Convolutional Neural Network (CNN). There is also pre-trained CNN models that are AlexNet and GoogLeNet, which produce excellent accuracy results.  ...  Deep Learning technique is gaining its popularity in computer vision and this paper applies this technique for face recognition problem.  ...  ACKNOWLEDGEMENTS The authors would like to thank Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA, Shah Alam, Selangor, for sponsoring this research.  ... 
doi:10.11591/ijeecs.v12.i2.pp476-481 fatcat:v7rmcg2wore2pb4sdr3ki4m7oa

Cutting the Error by Half: Investigation of Very Deep CNN and Advanced Training Strategies for Document Image Classification

Muhammad Zeshan Afzal, Andreas Kolsch, Sheraz Ahmed, Marcus Liwicki
2017 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR)  
The contribution of the paper is threefold: First, it investigates recently introduced very deep neural network architectures (GoogLeNet, VGG, ResNet) using transfer learning (from real images).  ...  Existing approaches, such as the DeepDocClassifier, apply standard Convolutional Network architectures with transfer learning from the object recognition domain.  ...  Although now we have a large dataset available for training document images, there is no study that shows the performance of very deep networks for large datasets of document images.  ... 
doi:10.1109/icdar.2017.149 dblp:conf/icdar/AfzalKAL17 fatcat:ob6v7obm6fhxbjixjlftb4jgfy

Object-Scene Convolutional Neural Networks for event recognition in images

Limin Wang, Zhe Wang, Wenbin Du, Yu Qiao
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Deep learning has turned out to be very effective in the task of object and scene recognition. For the tasks of object and recognition, there are very large-scale datasets: ImageNet and Places.  ...  Learning deep features for scene recognition using places database, NIPS, 2014. Figure : : The architecture of Object-Scene Convolutional Neural Network (OS-CNN) for event recognition.  ... 
doi:10.1109/cvprw.2015.7301333 dblp:conf/cvpr/WangWD015a fatcat:gim7lcqhhjc4pheeqwtnpr6dtu

Research on Substation Real-Time Object Recognition Algorithm Based on Deep Learning

2018 DEStech Transactions on Engineering and Technology Research  
The more advanced system based on the tensorflow framework is proposed for deep neural network recognition of objects in this paper.  ...  Our work is mainly the following: (1) Using the vgg deep convolutional network architecture as the infrastructure. (2) Real-time recognition of objects in video. (3) The system framework can be used in  ...  It is helpful for observing complex network structures and monitoring long-time and large-scale training.  ... 
doi:10.12783/dtetr/pmsms2018/24908 fatcat:36tpm3ddfrfcxiqkyqgximmjg4

Deep Learning Features at Scale for Visual Place Recognition [article]

Zetao Chen, Adam Jacobson, Niko Sunderhauf, Ben Upcroft, Lingqiao Liu, Chunhua Shen, Ian Reid, Michael Milford
2017 arXiv   pre-print
In this paper, we train, at large scale, two CNN architectures for the specific place recognition task and employ a multi-scale feature encoding method to generate condition- and viewpoint-invariant features  ...  The success of deep learning techniques in the computer vision domain has triggered a range of initial investigations into their utility for visual place recognition, all using generic features from networks  ...  In this paper we address these issues, presenting several advances towards the training of deep networks specifically for place recognition performance, at scale.  ... 
arXiv:1701.05105v1 fatcat:fww3cyeavrbvhm47iidqywlghy

Towards Good Practices for Very Deep Two-Stream ConvNets [article]

Limin Wang, Yuanjun Xiong, Zhe Wang, Yu Qiao
2015 arXiv   pre-print
Deep convolutional networks have achieved great success for object recognition in still images.  ...  However, for action recognition in videos, the improvement of deep convolutional networks is not so evident. We argue that there are two reasons that could probably explain this result.  ...  It is essentially a deep convolutional network architecture codenamed Inception, whose basic idea is Hebbian principle and the intuition of multi-scale processing.  ... 
arXiv:1507.02159v1 fatcat:wbczbtp7xjhyjbpxq26c7vh62m

Intelligent Object Recognition of Urban Water Bodies Based on Deep Learning for Multi-Source and Multi-Temporal High Spatial Resolution Remote Sensing Imagery

Shiran Song, Jianhua Liu, Yuan Liu, Guoqiang Feng, Hui Han, Yuan Yao, Mingyi Du
2020 Sensors  
A special convolutional neural network (CNN) based region proposal generation and object detection integrated framework has greatly improved the performance of object detection for HSRRSI, which provides  ...  This study uses the excellent "self-learning ability" of deep learning to construct a modified structure of the Mask R-CNN method which integrates bottom-up and top-down processes for water recognition  ...  However, due to the time-consuming and inefficient calculation of feature pyramids, the network usually extracts a single scale from a convolution layer of the underlying network, thus losing a large amount  ... 
doi:10.3390/s20020397 pmid:31936791 pmcid:PMC7014233 fatcat:37n5fijasjfzzbzd5djspbdr34
« Previous Showing results 1 — 15 out of 64,505 results