scholarly journals PornNet: A Unified Deep Architecture for Pornographic Video Recognition

2021 ◽  
Vol 11 (7) ◽  
pp. 3066
Author(s):  
Zhikang Fu ◽  
Jun Li ◽  
Guoqing Chen ◽  
Tianbao Yu ◽  
Tiansheng Deng
Keyword(s):  
Pattern Discrimination ◽  
Local Context ◽  
Context Aware ◽  
Large Dataset ◽  
Multimedia Resources ◽  
Deep Architecture ◽  
Recognition Result ◽  
Video Recognition ◽  
Image Context ◽  
Mental And Physical Health

In the era of big data, massive harmful multimedia resources publicly available on the Internet greatly threaten children and adolescents. In particular, recognizing pornographic videos is of great importance for protecting the mental and physical health of the underage. In contrast to the conventional methods which are only built on image classifier without considering audio clues in the video, we propose a unified deep architecture termed PornNet integrating dual sub-networks for pornographic video recognition. More specifically, with image frames and audio clues extracted from the pornographic videos from scratch, they are respectively delivered to two deep networks for pattern discrimination. For discriminating pornographic frames, we propose a local-context aware network that takes into account the image context in capturing the key contents, whilst leveraging an attention network which can capture temporal information for recognizing pornographic audios. Thus, we incorporate the recognition scores generated from the two sub-networks into a unified deep architecture, while making use of a pre-defined aggregation function to produce the whole video recognition result. The experiments on our newly-collected large dataset demonstrate that our proposed method exhibits a promising performance, achieving an accuracy at 93.4% on the dataset including 1 k pornographic samples along with 1 k normal videos and 1 k sexy videos.

Model and Infrastructure for Communications in Context-Aware Services

2011 ◽  
pp. 472-486
Author(s):  
Cristina Rodriguez-Sanchez ◽  
Susana Borromeo ◽  
Juan Hernandez-Tamames
Keyword(s):  
Mobile Communications ◽  
Local Environment ◽  
Local Context ◽  
Context Aware ◽  
The Novel ◽  
Novel Approach ◽  
Complex Process ◽  
Wide Range ◽  
Ambient Intelligent ◽  
New Type

The appearance of concepts such as “Ambient Intelligent”, “Ubiquitous Computing” and “Context-Awareness” is causing the development of a new type of services called “Context-Aware Services” that in turn may affect users of mobile communications. This technology revolution is a a complex process because of the heterogeneity of contents, devices, objects, technologies, resources and users that can coexist at the same local environment. The novel approach of our work is the development of a ”Local Infrastructure” in order to provide intelligent, transparent and adaptable services to the user as well as to solve the problem of local context control. Two contributions will be presented: conceptual model for developing a local infrastructure and an architecture design to control the service offered by the local infrastructure. This infrastructure proposed consists of an intelligent device network to link the personal portable device with the contextual services. The device design is modular, flexible, scalable, adaptable and reconfigurable remotely in order to tolerate new demanding services whenever are needed. Finally, the result suggests that we will be able to develop a wide range of new and useful applications, not conceived at origin.

scholarly journals An IoT Platform with Monitoring Robot Applying CNN-Based Context-Aware Learning

Sensors ◽  
2019 ◽  
Vol 19 (11) ◽  
pp. 2525 ◽  
Author(s):  
Moonsun Shin ◽  
Woojin Paik ◽  
Byungcheol Kim ◽  
Seonmin Hwang
Keyword(s):  
Machine Learning ◽  
High Performance ◽  
Contextual Information ◽  
Smart Phone ◽  
Context Aware ◽  
Context Processing ◽  
Intelligent Surveillance ◽  
Iot Platform ◽  
Fixed Type ◽  
Image Context

Internet of Things (IoT) technology has been attracted lots of interests over the recent years, due to its applicability across the various domains. In particular, an IoT-based robot with artificial intelligence may be utilized in various fields of surveillance. In this paper, we propose an IoT platform with an intelligent surveillance robot using machine learning in order to overcome the limitations of the existing closed-circuit television (CCTV) which is installed fixed type. The IoT platform with a surveillance robot provides the smart monitoring as a role of active CCTV. The intelligent surveillance robot, which has been built with its own IoT server, and can carry out line tracing and acquire contextual information through the sensors to detect abnormal status in an environment. In addition, photos taken by its camera can be compared with stored images of normal state. If an abnormal status is detected, the manager receives an alarm via a smart phone. For user convenience, the client is provided with an app to control the robot remotely. In the case of image context processing it is useful to apply convolutional neural network (CNN)-based machine learning (ML), which is introduced for the precise detection and recognition of images or patterns, and from which can be expected a high performance of recognition. We designed the CNN model to support contextually-aware services of the IoT platform and to perform experiments for learning accuracy of the designed CNN model using dataset of images acquired from the robot. Experimental results showed that the accuracy of learning is over 0.98, which means that we achieved enhanced learning in image context recognition. The contribution of this paper is not only to implement an IoT platform with active CCTV robot but also to construct a CNN model for image-and-context-aware learning and intelligence enhancement of the proposed IoT platform. The proposed IoT platform, with an intelligent surveillance robot using machine learning, can be used to detect abnormal status in various industrial fields such as factory, smart farms, logistics warehouses, and public places.

scholarly journals Retrieval of Flower Videos Based on a Query With Multiple Species of Flowers

2021 ◽  
Author(s):  
Manjunath Aradhya ◽  
Jyothi VK ◽  
Sharath Kumar ◽  
Guru DS
Keyword(s):  
Machine Learning ◽  
Single Species ◽  
Video Data ◽  
Support Vector ◽  
Large Collection ◽  
Large Dataset ◽  
Linear Discriminant ◽  
Video Recognition ◽  
Multiclass Support Vector Machine ◽  
Multiple Species

Searching, recognizing and retrieving a video of interest from a large collection of a video data is an instantaneous requirement. This requirement has been recognized as an active area of research in computer vision, machine learning and pattern recognition. Flower video recognition and retrieval is vital in the field of floriculture and horticulture. In this paper we propose a model for the retrieval of videos of flowers. Initially, videos are represented with keyframes and flowers in keyframes are segmented from their background. Then, the model is analysed by features extracted from flower regions of the keyframe. A Linear Discriminant Analysis (LDA) is adapted for the extraction of discriminating features. Multiclass Support Vector Machine (MSVM) classifier is applied to identify the class of the query video. Experiments have been conducted on relatively large dataset of our own, consisting of 7788 videos of 30 different species of flowers captured from three different devices. Generally, retrieval of flower videos is addressed by the use of a query video consisting of a flower of a single species. In this work we made an attempt to develop a system consisting of retrieval of similar videos for a query video consisting of flowers of different species.

scholarly journals When learning Italian as a Second Language, tourism and technology go hand in hand

2021 ◽  
Author(s):  
Anna Zingaro ◽  
Cristiana Cervini
Keyword(s):  
Language Learning ◽  
Mobile Devices ◽  
Mobile Application ◽  
Incidental Learning ◽  
Local Context ◽  
Learning Needs ◽  
Context Aware ◽  
Direct Experience ◽  
Language And Culture ◽  
The City

This paper aims to describe the development of CALL-ER, an application for mobile devices, produced within the CALL-ER project (Context-Aware Language Learning in Emilia Romagna). An ever-increasing availability of applications for language learning that meet the different learning needs of users, as well as the ubiquitous wireless communication, led applications for mobile devices to become gradually more context-aware. This means that language is acquired by users through the direct experience with the local context where they are. An example in this regard is represented by the CALL-ER mobile application, that supports mobility students through the incidental learning of Italian language and culture in the city of Forlì. We will begin this contribution with an outline of the theoretical underpinnings that supported the project and a presentation of the project itself. We will then present the first stage of the project, during which the application was developed before its first testing. At this point, an overall description of the application will be given. A special attention will be paid throughout this paper both to how language learning has been conceived through experiential tourism and to the multimodality of the contents.

Adult Image Classification by a Local-Context Aware Network

2018 ◽  
Author(s):  
Xizi Wang ◽  
Feng Cheng ◽  
Shilin Wang ◽  
Huanrong Sun ◽  
Gongshen Liu ◽  
...  
Keyword(s):  
Image Classification ◽  
Local Context ◽  
Context Aware

scholarly journals A Residual Attention and Local Context-Aware Network for Road Extraction from High-Resolution Remote Sensing Imagery

2021 ◽  
Vol 13 (24) ◽  
pp. 4958
Author(s):  
Ziwei Liu ◽  
Mingchang Wang ◽  
Fengyan Wang ◽  
Xue Ji
Keyword(s):  
Remote Sensing ◽  
High Resolution ◽  
Spectral Characteristics ◽  
Attention Mechanism ◽  
Image Resolution ◽  
Feature Representation ◽  
Local Context ◽  
Context Aware ◽  
Multi Scale ◽  
Feature Information

Extracting road information from high-resolution remote sensing images (HRI) can provide crucial geographic information for many applications. With the improvement of remote sensing image resolution, the image data contain more abundant feature information. However, this phenomenon also enhances the spatial heterogeneity between different types of roads, making it difficult to accurately discern the road and non-road regions using only spectral characteristics. To remedy the above issues, a novel residual attention and local context-aware network (RALC-Net) is proposed for extracting a complete and continuous road network from HRI. RALC-Net utilizes a dual-encoder structure to improve the feature extraction capability of the network, whose two different branches take different feature information as input data. Specifically, we construct the residual attention module using the residual connection that can integrate spatial context information and the attention mechanism, highlighting local semantics to extract local feature information of roads. The residual attention module combines the characteristics of both the residual connection and the attention mechanism to retain complete road edge information, highlight essential semantics, and enhance the generalization capability of the network model. In addition, the multi-scale dilated convolution module is used to extract multi-scale spatial receptive fields to improve the model’s performance further. We perform experiments to verify the performance of each component of RALC-Net through the ablation study. By combining low-level features with high-level semantics, we extract road information and make comparisons with other state-of-the-art models. The experimental results show that the proposed RALC-Net has excellent feature representation ability and robust generalizability, and can extract complete road information from a complex environment.

scholarly journals Multi-Perspective Cost-Sensitive Context-Aware Multi-Instance Sparse Coding and Its Application to Sensitive Video Recognition

2016 ◽  
Vol 18 (1) ◽  
pp. 76-89 ◽  
Author(s):  
Weiming Hu ◽  
Xinmiao Ding ◽  
Bing Li ◽  
Jianchao Wang ◽  
Yan Gao ◽  
...  
Keyword(s):  
Sparse Coding ◽  
Context Aware ◽  
Video Recognition

scholarly journals ENSINANDO A DIVERSIDADE OU A TRANSFOBIA? UM PANORAMA DA EDUCAÇÃO SOBRE DIVERSIDADE SEXUAL E DE GÊNERO NAS ESCOLAS DA REGIÃO DE SOROCABA-SP E SUA INTERSECÇÃO COM SAÚDE MENTAL

2020 ◽  
Vol 1 ◽  
pp. 173
Author(s):  
Sara Laham Sonetti ◽  
Marcos Roberto Vieira Garcia
Keyword(s):  
Mental Health ◽  
Protective Role ◽  
Local Context ◽  
Sexuality And Gender ◽  
Prejudice And Discrimination ◽  
And Gender ◽  
Relations Of Power ◽  
Forms Of Violence ◽  
Mental And Physical Health

A estrutura social e os padrões de normalidade refletem diretamente as relações de poder e a naturalização de algumas normas em detrimento de outras. Temos no Brasil uma heteronormatividade compulsória vigente, que se apresenta também dentro da escola, nos conteúdos ensinados e nas condutas de comportamento induzidas ou exigidas. O presente artigo discute esse tema no panorama nacional de forma mais ampla e na região de Sorocaba, em particular. Nesse contexto local, a heteronormatividade tem sido afirmada e reafirmada através de aprovações de leis que desconsideram a identidade de gênero e de medidas que reforçam o preconceito e discriminação em torno de diversas formas de expressão da sexualidade e gênero. Esse desrespeito e as diferentes formas de violências dele advindas são desfavoráveis à saúde mental e física de pessoas que não se enquadram na cisheteronormatividade, fazendo então da escola um ambiente potencialmente lesivo a alunos e funcionários LGBT. Ao mesmo tempo, há a possibilidade de a escola exercer um papel protetivo, ao promover o debate e educação sobre sexualidade, o que tem sido pauta de movimentos sociais sorocabanos ao reivindicarem mudanças nas leis e diminuição da influência do conservadorismo presente no meio político local. Palavras-chave: Travestis. Transexuais. Heteronormatividade. Saúde Mental. Transfobia na escola.ABSTRACTThe social structure and patterns of normality reflect directly the relations of power and naturalization of some norms in detriment of others. There is a prevailing compulsory heteronormativity, which also occurs into the school, within the limits taught and conducts of induced or required behaviors. The present article discusses the theme in the national panorama in a broader way and in the region of Sorocaba, in particular way. In this local context, the heteronormativity has been affirmed and reaffirmed through approvals of laws that disregard gender identity and policies that reinforce prejudice and discrimination in the senses of the expressions of sexuality and gender. This disrespect and the different forms of violence that come from it, are unfavorable to mental and physical health of people that doesn’t fit in the cisheteronormativity, making the school to become a harmfull place to LGBT studants and employers. At the same time, there is a possibility of a protective role of school, while promoting debate and education about sexuality, which has been some of the schedule of social moviments of Sorocaba, that claim for chances in the law and for decrease of the influence of conservatism in the local political environment.Keywords: Transvestite. Transsexuals. Heteronormativity. Mental Health. Transphobia into schools.

scholarly journals Context-Aware and Locality-Constrained Coding for Image Categorization

2014 ◽  
Vol 2014 ◽  
pp. 1-14
Author(s):  
Wenhua Xiao ◽  
Bin Wang ◽  
Yu Liu ◽  
Weidong Bao ◽  
Maojun Zhang
Keyword(s):  
Feature Space ◽  
Local Features ◽  
Local Context ◽  
Context Information ◽  
Context Aware ◽  
Image Categorization ◽  
Constrained Coding ◽  
The Arts ◽  
Coding Strategy ◽  
Further Development

Improving the coding strategy for BOF (Bag-of-Features) based feature design has drawn increasing attention in recent image categorization works. However, the ambiguity in coding procedure still impedes its further development. In this paper, we introduce a context-aware and locality-constrained Coding (CALC) approach with context information for describing objects in a discriminative way. It is generally achieved by learning a word-to-word cooccurrence prior to imposing context information over locality-constrained coding. Firstly, the local context of each category is evaluated by learning a word-to-word cooccurrence matrix representing the spatial distribution of local features in neighbor region. Then, the learned cooccurrence matrix is used for measuring the context distance between local features and code words. Finally, a coding strategy simultaneously considers locality in feature space and context space, while introducing the weight of feature is proposed. This novel coding strategy not only semantically preserves the information in coding, but also has the ability to alleviate the noise distortion of each class. Extensive experiments on several available datasets (Scene-15, Caltech101, and Caltech256) are conducted to validate the superiority of our algorithm by comparing it with baselines and recent published methods. Experimental results show that our method significantly improves the performance of baselines and achieves comparable and even better performance with the state of the arts.

Sign in / Sign up

Export Citation Format

Share Document