Smart Content Recognition from Images Using a Mixture of Convolutional Neural Networks

Connie, Tee; Al-Shabi, Mundher; Goh, Michael Kah Ong

doi:10.1007/978-981-10-6451-7_2

Cited by 12 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, we use the following three datasets: Adult content dataset, NudeNet, and LPSD (Table 1). Adult content dataset [16] is one of the earliest datasets in this field and contains two categories: 'safe' and 'adult'. NudeNet dataset [14] includes an intermediate label 'sexy', in addition to the 'safe' and 'nude' labels.…”

Section: Nudity Classification Datasetsmentioning

confidence: 99%

State-of-the-Art in Nudity Classification: A Comparative Analysis

Akyon

Temizel

2023

2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)

View full text Add to dashboard Cite

This paper presents a comparative analysis of existing nudity classification techniques for classifying images based on the presence of nudity, with a focus on their application in content moderation. The evaluation focuses on CNN-based models, vision transformer, and popular opensource safety checkers from Stable Diffusion and Largescale Artificial Intelligence Open Network (LAION). The study identifies the limitations of current evaluation datasets and highlights the need for more diverse and challenging datasets. The paper discusses the potential implications of these findings for developing more accurate and effective image classification systems on online platforms. Overall, the study emphasizes the importance of continually improving image classification models to ensure the safety and well-being of platform users. The project page, including the demonstrations and results is publicly available at https://github.com/fcakyon/content-moderation-deep-learning.

show abstract

Section: Nudity Classification Datasetsmentioning

confidence: 99%

State-of-the-Art in Nudity Classification: A Comparative Analysis

Akyon

Temizel

2023

2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)

View full text Add to dashboard Cite

show abstract

“…Consequently, these datasets cannot be used to benchmark other pornography detection methods. Connie [6] built a dataset for adult content recognition in images. This dataset, consisting of 41,154 pornographic images and 40,152 neutral images, is open but its images are of fixed size (128 x 128 pixels); therefore, they are generally unsuitable for fair evaluations.…”

Section: Pornography Datasetmentioning

confidence: 99%

“…Pornographic image classification requires a dataset of images divided into categories, while object detection and segmentation require object annotations. Although some datasets for pornography classification are available [4][5][6], their image quality is too low for today's applications. These datasets haven't been updated for a while, which makes them incapable to recognize recent types of pornography.…”

Section: Introductionmentioning

confidence: 99%

LSPD: A Large-Scale Pornographic Dataset for Detection and Classification

2022

IJIES

View full text Add to dashboard Cite

This paper introduces a new dataset named Large-Scale Pornographic Dataset for detection and classification (LSPD) that intends to advance the standard quality of pornographic visual content classification and sexual object detection tasks. As we recognize, the LSPD dataset is the first ever dataset for both object detection and image/video classification tasks in this area. The dataset gathers a large-scale corpus of pornographic/nonpornographic images and videos containing a rich diversity of context. The images and videos are not only labelled with their representative class but are also annotated by polygon masks of four private sexual objects (breasts, male and female genitals, and anuses). Our dataset contains 500,000 images and 4,000 videos, with more than 50,000 annotated images. To ensure fair use of the dataset, we present a detailed statistical analysis and provide baseline benchmarking scenarios for both image/video classification and instance detection/segmentation tasks. Finally, we evaluate the performance of four object detection algorithms and a Convolutional Neural Network (CNN) classifier on these scenarios.

show abstract

“…Ensemble methods were also utilised in adult content recognition [38][39][40]. Here, a weighted sum of several deep neural networks (DNNs) was used to express the CNN's weights as a linear regression problem learned using ordinary least squares (OLS) [38]. Additionally, ensemble framework uncertain inference employed a Bayesian network.…”

Section: Introductionmentioning

confidence: 99%

Transfer Detection of YOLO to Focus CNN’s Attention on Nude Regions for Adult Content Detection

et al. 2020

View full text Add to dashboard Cite

Video pornography and nudity detection aim to detect and classify people in videos into nude or normal for censorship purposes. Recent literature has demonstrated pornography detection utilising the convolutional neural network (CNN) to extract features directly from the whole frames and support vector machine (SVM) to classify the extracted features into two categories. However, existing methods were not able to detect the small-scale content of pornography and nudity in frames with diverse backgrounds. This limitation has led to a high false-negative rate (FNR) and misclassification of nude frames as normal ones. In order to address this matter, this paper explores the limitation of the existing convolutional-only approaches focusing the visual attention of CNN on the expected nude regions inside the frames to reduce the FNR. The You Only Look Once (YOLO) object detector was transferred to the pornography and nudity detection application to detect persons as regions of interest (ROIs), which were applied to CNN and SVM for nude/normal classification. Several experiments were conducted to compare the performance of various CNNs and classifiers using our proposed dataset. It was found that ResNet101 with random forest outperformed other models concerning the F1-score of 90.03% and accuracy of 87.75%. Furthermore, an ablation study was performed to demonstrate the impact of adding the YOLO before the CNN. YOLO–CNN was shown to outperform CNN-only in terms of accuracy, which was increased from 85.5% to 89.5%. Additionally, a new benchmark dataset with challenging content, including various human sizes and backgrounds, was proposed.

show abstract

Smart Content Recognition from Images Using a Mixture of Convolutional Neural Networks

Cited by 12 publications

References 11 publications

State-of-the-Art in Nudity Classification: A Comparative Analysis

State-of-the-Art in Nudity Classification: A Comparative Analysis

LSPD: A Large-Scale Pornographic Dataset for Detection and Classification

Transfer Detection of YOLO to Focus CNN’s Attention on Nude Regions for Adult Content Detection

Contact Info

Product

Resources

About