AcFR: Active Face Recognition Using Convolutional Neural Networks

Nakada, Masaki; Wang, Han; Terzopoulos, Demetri

doi:10.1109/cvprw.2017.11

Cited by 32 publications

(18 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A classification stage, which is often a shallow architecture such as a multilayer perceptron (MLP), follows these stages of feature extraction. Characterized by a relatively low memory footprint, CNN have witnessed tremendous success in image classification [1], localization and detection [22], speech recognition [23] and a wide range of visual processing tasks [24].…”

Section: B Computational Models Of the Visual Cortexmentioning

confidence: 99%

On Biologically Inspired Stochastic Reinforcement Deep Learning: A Case Study on Visual Surveillance

Hajj

Awad

2019

IEEE Access

View full text Add to dashboard Cite

Here, we present a biologically inspired visual network (BIVnet) for image processing tasks. The proposed model possesses similarities with its neural counterpart and is trained by a stochastic algorithm which employs a partially observable Markov decision process to execute a reinforcement learning strategy. The network was tested on a collection of available datasets in surveillance-related tasks and showed superior performance compared with the state-of-the-art architectures. An average improvement of 15.2% in accuracy on a collection of publicly available image datasets is shown in our experimental results. INDEX TERMSAbandoned luggage detection, deep learning, person identification, visual tasks.

show abstract

Section: B Computational Models Of the Visual Cortexmentioning

confidence: 99%

On Biologically Inspired Stochastic Reinforcement Deep Learning: A Case Study on Visual Surveillance

Hajj

Awad

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Meanwhile, it is also impeded by the massive data. Deep Learning is widely applied to various fields of computer vision [ 22 ], vehicle detection [ 23 ], and gesture recognition [ 24 ] and so on. Excitingly, it is the first time that Deep Learning is introduced to the recognition of the point symbols in topographic maps.…”

Section: Related Workmentioning

confidence: 99%

A Combinatorial Solution to Point Symbol Recognition

Quan

Shi

Miao

et al. 2018

Sensors

View full text Add to dashboard Cite

Recent work has shown that recognizing point symbols is an essential task in the field of map digitization. For the identification of symbols, it is generally necessary to compare the symbols with a specific criterion and find the most similar one with each known symbol one by one. Most of the works can only identify a single symbol, a small number of works are to deal with multiple symbols simultaneously with a low recognition accuracy. Given the two deficiencies, this paper proposes a deep transfer learning architecture, where the task is to learn a symbol classifier with AlexNet. For the insufficient dataset, we develop a method for transfer learning that uses a MNIST dataset to pretrain the model, which makes up for the problem of small training dataset and enhances the generalization of the model. Before the recognition process, preprocessing the point symbols in the map to coarse screening out the areas suspected of point symbols. We show a significant improvement over using point symbol images to keep a high performance in being able to deal with many more categories of symbols simultaneously.

show abstract

“…The asymptomatic cases of COVID-19 pneumonia CT scan images have definite characteristics and have important value in screening and detecting patients with COVID-19 pneumonia, especially in the highly suspicious cases with negative PCR testing. [ 7 , 8 ]. Another diagnostic method to diagnosis and treatment guideline for COVID-19 pneumonia, issued by the National Health Commission of The People's Republic of China is the CT, especially the high-resolution CT (HRCT) [ 9 ].…”

Section: Introductionmentioning

confidence: 99%

“…The CNN designs are based on the architecture that performs traffic sign classification related tasks with a good recognition rate. This method uses a deep CNN which has been widely used for image classification and trained to classify faces using a VGG face (Deep Face) Figure 1 [ 8 , 22 ]. The CNN “deep learning,” in medical imaging is an explosively growing, promising field [ 23 , 24 ].…”

Section: Introductionmentioning

confidence: 99%

A Study of a New Technique of the CT Scan View and Disease Classification Protocol Based on Level Challenges in Cases of Coronavirus Disease

Salamh

Salamah

Akyüz

2021

Radiology Research and Practice

View full text Add to dashboard Cite

The chest Computer Tomography (CT scan) is used in the diagnosis of coronavirus disease 2019 (COVID-19) and is an important complement to the Reverse Transcription Polymerase Chain Reaction (RT-PCR) test. The paper aims to improve the radiological diagnosis in the case of coronavirus disease COVID-19 pneumonia on forms of noninvasive approaches with conventional and high-resolution computer tomography (HRCT) scan images upon chest CT images of patients confirmed with mild to severe findings. The preliminary study is to compare the radiological findings of COVID-19 pneumonia in conventional chest CT images with images processed by a new tool and reviewed by expert radiologists. The researchers used a new filter called Golden Key Tool (GK-Tool) which has confirmed the improvement in the quality and diagnostic efficacy of images acquired using our modified images. Further, Convolution Neural Networks (CNNs) architecture called VGG face was used to classify chest CT images. The classification has been performed by using VGG face on various datasets which are considered as a protocol to diagnose COVID-19, Non-COVID-19 (other lung diseases), and normal cases (no findings on chest CT). Accordingly, the performance evaluation of the GK-Tool was fairly good as shown in the first set of results, where 80–95% of participants show a good to excellent assessment of the new images view in the case of COVID-19 patients. The results, in general, illustrate good recognition rates in the diagnosis and, therefore, would be significantly higher in normal cases with COVID-19. These results could reduce the radiologist’s workload burden and play a major role in the decision-making process.

show abstract

AcFR: Active Face Recognition Using Convolutional Neural Networks

Cited by 32 publications

References 12 publications

On Biologically Inspired Stochastic Reinforcement Deep Learning: A Case Study on Visual Surveillance

On Biologically Inspired Stochastic Reinforcement Deep Learning: A Case Study on Visual Surveillance

A Combinatorial Solution to Point Symbol Recognition

A Study of a New Technique of the CT Scan View and Disease Classification Protocol Based on Level Challenges in Cases of Coronavirus Disease

Contact Info

Product

Resources

About