Analyzing periodicity and saliency for adult video detection

Abstract

Content-based adult video detection plays an important role in preventing pornography.However, existing methods usually rely on single modality and seldom focus on multi-modality semantics representation. Addressing at this problem, we put forward an approach of analyzing periodicity and saliency for adult video detection. At first, periodic patterns and salient regions are respectively analyzed in audio-frames and visual-frames. Next, the multi-modal co-occurrence semantics is described by combining audio periodicity with visual saliency. Moreover, the performance of our approach is evaluated step by step. Experimental results show that our approach obviously outperforms some state-of-the-art methods.

Obscene image detection using transfer learning and feature fusion

“…BossaNova integrates both color information and shape description hence outperformed the conventional BoWs based method. In [ 18 ], authors fused audio codebook algorithm and visual codebook algorithm by the SURF technique to preform the detection process. Appropriate selection of codebook size, pooling algorithm and threshold parameter play vital role in defining accuracy of these methods.…”

“…Analyzing Tables 1 and 2 few inferences are drawn: BoWs based method in [ 18 ] by utilizing audio, video code book algorithm and SURF feature descriptor provides detection accuracy of 96 % when tested over a limited number (50) of videos at a cost of heavy computational overload. As compared to conventional methods, CNN based methods outperform in terms of detection accuracy and computational time.…”

“…In literature, there exists varieties of obscene image detection methodologies (OIDMs) for identification and detection of obscene contents and consequently enable to prohibit/block their further use at the user end. These methodologies are broadly categorized as i) skin-color segmentation based methods [ 5 , 8 , 16 ] ii) Bag of words (BoWs) based models [ 6 , 14 , 18 ] iii) visual and motion feature analysis based [ 11 , 19 , 21 , 30 , 31 ] and iv) deep learning-based classification strategies [ 1 , 4 , 7 , 9 , 22 , 24 – 27 , 29 , 32 , 33 ]. Usually, skin-color segmentation based methods are highly dependent on human skin pixels and hence, unable to generalize the findings.…”

Bare Skin Image Classification using Convolution Neural Network

“…In [27] Convolutional Neural Network using deep learning was used for detection and classification of objects related to metal polishing shaft surface. In [28], analysis were carried out to periodicity and saliency for adult video detection. Pornographic images and bikini images were classified using machine learning based forensic discriminator [29].…”

A Deep Learning Approach to Detect Pornography Videos in Educational Repositories

“…al. [Liu et al 2020] propose a multimodal approach to pornography detection, it uses audio-frames and visual-frames to create handcrafted low-level features based on, respectively on periodic patterns and salient regions. Once those features are extracted, they use k-means clustering to create audio and visual codebooks.…”