Robust face detection using local CNN and SVM based on kernel combination

Qing, Tao; Shu, Zhan; Xiao-hong, LI; Kurihara, Toru

doi:10.1016/j.neucom.2015.10.139

Cited by 65 publications

(25 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the article [6], the authors have used this technique for detection of vehicle logo. An alternate way of object detection is deep neural networks, as [7] uses CNN and SVM for face detection. The authors have shown that CNN and SVM give better results than other popular methods for face recognition.…”

Section: Existing Workmentioning

confidence: 99%

An application of pre-trained CNN for image classification

Abdullah

Hasan

2017

2017 20th International Conference of Computer and Information Technology (ICCIT)

View full text Add to dashboard Cite

One of the important practical applications of object detection and image classification can be for security enhancement. If dangerous objects e.g. knives can be identified automatically, then a lot of violence can be prevented. For this purpose, various different algorithms and methods are out there that can be used. In this paper, four of them have been investigated to find out which can identify knives from a dataset of images more accurately. Among Bag of Words, HOG-SVM, CNN and pre-trained Alexnet CNN, the deep learning CNN methods are found to give best results, though they consume significantly more resources.

show abstract

Section: Existing Workmentioning

confidence: 99%

An application of pre-trained CNN for image classification

Abdullah

Hasan

2017

2017 20th International Conference of Computer and Information Technology (ICCIT)

View full text Add to dashboard Cite

show abstract

Section: Existing Workmentioning

confidence: 99%

An analysis of Feature extraction and Classification Algorithms for Dangerous Object Detection

Kibria

Hasan

2017

2017 2nd International Conference on Electrical &Amp; Electronic Engineering (ICEEE)

View full text Add to dashboard Cite

Abstract-One of the important practical applications of object detection and image classification can be for security enhancement. If dangerous objects e.g. knives can be identified automatically, then a lot of violence can be prevented. For this purpose, various different algorithms and methods are out there that can be used. In this paper, four of them have been investigated to find out which can identify knives from a dataset of images more accurately. Among Bag of Words, HOG-SVM, CNN and pre-trained Alexnet CNN, the deep learning CNN methods are found to give best results, though they consume significantly more resources.

show abstract

“…The key challenge of face detection [15] is the large appearance variations due to the number of real-world factors such as pose changes, exaggerated expressions and extreme illuminations which can lead to the large intra-class variations and making the detection algorithm not robust enough. A locality sensitive support vector machine using kernel combination (LS-KC-SVM) algorithm to solve the above two problems is proposed.…”

Section: Convolution Neural Networkmentioning

confidence: 99%

Efficient Object Recognition using Convolution Neural Networks Theorem

Thakral¹,

Shekhar²,

Victor³

2017

IJCA

View full text Add to dashboard Cite

Object recognition is the process of identification of an object in an image. There exist various algorithms for the same. Appearance based algorithms have demonstrated good efficiency, however, their performance gets affected adversely in the presence of clutter or when background changes are affected. We hope to overcome this issue by using Convolution Neural Network (CNN) Theorem. The approach is shape based and has been proven to work well under broad range of circumstances: varied lighting conditions, affine transformations, etc. It involves tiling, which is the phenomenon of the use of multiple layers of neurons to process small portions of the image, which are then used to obtain better representations of the image. This allows CNN to be translation-tolerant. The neural elements learn to recognize objects about which they have no previous information, this "learning" mechanism is affected by the fact that representations of the image are learned by the inner layers of the deep architectures of neurons. Unlike RBM and Auto-encoder, which are capable of learning only single global weight matrix layers, the CNN theorem makes use of shared weight in convolution layers, which means that the same filter (weight bank) is used for each pixel in the layer, which reduces the memory footprint and improves performance.

show abstract

Robust face detection using local CNN and SVM based on kernel combination

Cited by 65 publications

References 47 publications

An application of pre-trained CNN for image classification

An application of pre-trained CNN for image classification

An analysis of Feature extraction and Classification Algorithms for Dangerous Object Detection

Efficient Object Recognition using Convolution Neural Networks Theorem

Contact Info

Product

Resources

About