Image Safeguarding: Reasoning with Conditional Vision Language Model and Obfuscating Unsafe Content Counterfactually
Open Access
- 24 March 2024
- journal article
- research article
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 38 (2) , 774-782
Abstract
Social media platforms are being increasingly used by malicious actors to share unsafe content, such as images depicting sexual activity, cyberbullying, and self-harm. Consequently, major platforms use artificial intelligence (AI) and human moderation to obfuscate such images to make them safer. Two critical needs for obfuscating unsafe images is that an accurate rationale for obfuscating image regions must be provided, and the sensitive regions should be obfuscated (e.g. blurring) for users' safety. This process involves addressing two key problems: (1) the reason for obfuscating unsafe images demands the platform to provide an accurate rationale that must be grounded in unsafe image-specific attributes, and (2) the unsafe regions in the image must be minimally obfuscated while still depicting the safe regions. In this work, we address these key issues by first performing visual reasoning by designing a visual reasoning model (VLM) conditioned on pre-trained unsafe image classifiers to provide an accurate rationale grounded in unsafe image attributes, and then proposing a counterfactual explanation algorithm that minimally identifies and obfuscates unsafe regions for safe viewing, by first utilizing an unsafe image classifier attribution matrix to guide segmentation for a more optimal subregion segmentation followed by an informed greedy search to determine the minimum number of subregions required to modify the classifier's output based on attribution score. Extensive experiments on uncurated data from social networks emphasize the efficacy of our proposed method. We make our code available at: https://github.com/SecureAIAutonomyLab/ConditionalVLMKeywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Explainable image classification with evidence counterfactualPattern Analysis and Applications, 2022
- Deplatforming sexual speech in the age of FOSTA/SESTAPorn Studies, 2021
- How Instagram’s algorithm is censoring women and vulnerable users but helping online abusersFeminist Media Studies, 2020
- A Practitioner Survey Exploring the Value of Forensic Tools, AI, Filtering, & Safer Presentation for Investigating Child Sexual Abuse Material (CSAM)Digital Investigation, 2019
- Self-Harm, Suicidal Behaviours, and Cyberbullying in Children and Young People: Systematic ReviewJournal of Medical Internet Research, 2018
- Grad-CAM++: Generalized Gradient-Based Visual Explanations for Deep Convolutional Networks2018
- Effectiveness and Users' Experience of Obfuscation as a Privacy-Enhancing Technology for Sharing PhotosProceedings of the ACM on Human-Computer Interaction, 2017
- SLIC Superpixels Compared to State-of-the-Art Superpixel MethodsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2012
- Identifying and Managing Stress in Child Pornography and Child Exploitation InvestigatorsJournal of Police and Criminal Psychology, 2009
- Efficient Graph-Based Image SegmentationInternational Journal of Computer Vision, 2004