Ferramenta PPCensor: detecção de pornografia em tempo real no streaming de vídeo

• Jackson Mallmann IFC / PUCPR
• Altair O. Santin PUCPR
• Eduardo K. Viegas PUCPR
• Roger R. dos Santos PUCPR
• Jhonatan Geremias PUCPR

DOI: https://doi.org/10.5753/sbseg_estendido.2020.19275

Abstract

This paper presents the tool entitled Private Parts Censor (PPCensor) for the detection of pornography-related objects in videos in a network proxy. To achieve such a goal, the tool performs the analysis of the video frames that are currently being downloaded in real time in a transparent manner. For the pornography object detection, an object detector is trained with PPO (Private Parts Object Dataset). When PPCensor identifies a video frame containing private parts (objects), its content is hidden from the current video without the need for user interaction or the requirement of additional processing in the user device. The evaluation results have shown that PPCensor is able to detect private parts in real time during video transmissions.

References

Lin, T.; Maire, M.; Belongie, S.; Bourdev, L.; Girshick, R.; Hays, J.; Perona, P.; Ramanan, D.; Zitnick, C.L.; Dollár, P. (2014). “Microsoft COCO: Common objects in context”. In ECCV, 2014.

Mallmann, J.; Santin, A.O.; Viegas, E.K.; Santos, R.R. DOS.; Geremias, J. (2020). “PPCensor: Architecture for real-time pornography detection in video streaming”. Future Generation Computer Systems, v. 112, p. 945-955, 2020.

Moreira, D.; Avila, S.; Perez, M.; Moraes, D.; Testoni, V.; Valle, E.; Goldenstein, S.; Rocha, A. (2016). “Pornography classification: The hidden clues in video spacetime”, Forensic Science Int. Vol. 268, p. 46-61.

Oquab, M.; Bottou, L.; Laptev, I.; Sivic, J. (2014). “Learning and transferring mid-level image representations using convolutional neural networks” in Proc. of the IEEE conf. on computer vision and pattern recognition, pp. 1717–1724.

Pafy. (2020), “Library to download YouTube content and retrieve metadata”, https://pythonhosted.org/pafy/.

PCAA. (2020), “Prevent Child Abuse America”, https://preventchildabuse.org/.

Pornhub. (2020), “2018 Year in Review”, https://www.pornhub.com/insights/.

Ren, S.; He, K.; Girshick, R.; Sun, J. (2017) "Faster r-cnn: towards real-time object detection with region proposal networks", IEEE Trans. Pattern Anal. Mach. Intell., vol. 39, no. 6, pp. 1137-1149.

Statista. (2020), “How Much of the Internet Consists of Porn?”, https://www.statista.com/chart/16959/share-of-the-internet-that-is-porn/.

Squid. (2020), “Squid Development Projects”. http://devel.squid-cache.org/.

TensorFlow. (2020), “Tensorflow Object Detection API”, https://www.tensorflow.org/.

Vicentini, C.; Santin, A.; Viegas, E.; Abreu, V. (2018) A machine learning auditing model for detection of multi-tenancy issues within tenant domain. IEEE International Symposium on Cluster, Cloud and Grid Computing, CCGRID, p. 543–552.

Viegas, E.; Santin, A. O.; Franca, A.; et al (2017). Towards an energy-efficient anomaly-based intrusion detection engine for embedded systems. IEEE Transactions on Computers, v. 66, n. 1, 2017.