Ayan Sar, Sampurna Roy, Tanupriya Choudhury, Ajith Abraham
Generative adversarial networks (GANs) and diffusion models have dramatically advanced deepfake technology, and its threats to digital security, media integrity, and public trust have increased rapidly. This research explored zero-shot deepfake detection, an emerging method even when the models have never seen a particular deepfake variation. In this work, we studied self-supervised learning, transformer-based zero-shot classifier, generative model fingerprinting, and meta-learning techniques that better adapt to the ever-evolving deepfake threat. In addition, we suggested AI-driven prevention strategies that mitigated the underlying generation pipeline of the deepfakes before they occurred. They consisted of adversarial perturbations for creating deepfake generators, digital watermarking for content authenticity verification, real-time AI monitoring for content creation pipelines, and blockchain-based content verification frameworks. Despite these advancements, zero-shot detection and prevention faced critical challenges such as adversarial attacks, scalability constraints, ethical dilemmas, and the absence of standardized evaluation benchmarks. These limitations were addressed by discussing future research directions on explainable AI for deepfake detection, multimodal fusion based on image, audio, and text analysis, quantum AI for enhanced security, and federated learning for privacy-preserving deepfake detection. This further highlighted the need for an integrated defense framework for digital authenticity that utilized zero-shot learning in combination with preventive deepfake mechanisms. Finally, we highlighted the important role of interdisciplinary collaboration between AI researchers, cybersecurity experts, and policymakers to create resilient defenses against the rising tide of deepfake attacks.
Visual homing is a lightweight approach to visual navigation. Given the stored information of an initial 'home' location, the navigation task back to this location is achieved from any other location by comparing the stored home information to the current image and extracting a motion vector. A challenge that constrains the applicability of visual homing is that the home location must be within the robot's field of view to initiate the homing process. Thus, we propose a blockchain approach to visual navigation for a heterogeneous robot team over a wide area of visual navigation. Because it does not require map data structures, the approach is useful for robot platforms with a small computational footprint, and because it leverages current visual information, it supports a resilient and adaptive path selection. Further, we present a lightweight Proof-of-Work (PoW) mechanism for reaching consensus in the untrustworthy visual homing network.
The Ethereum blockchain permits the development and deployment of smart contracts which can store and execute code 'on-chain' - that is, entirely on nodes in the blockchain's network. Smart contracts have traditionally been used for financial purposes, but since smart contracts are Turing-complete, their algorithmic scope is broader than any single domain. To that end, we design, develop, and deploy a comprehensive 3D rendering engine programmed entirely in Ethereum smart contracts, called Shackled. Shackled computes a 2D image from a 3D scene, executing every single computation on-chain, on Ethereum. To our knowledge, Shackled is the first and only fully on-chain 3D rendering engine for Ethereum. In this work, we 1) provide three unique datasets for the purpose of using and benchmarking Shackled, 2) execute said benchmarks and provide results, 3) demonstrate a potential use case of Shackled in the domain of tokenised generative art, 4) provide a no-code user interface to Shackled, 5) enumerate the challenges associated with programming complex algorithms in Solidity smart contracts, and 6) outline potential directions for improving the Shackled platform. It is our hope that this work increases the Ethereum blockchain's native graphics processing capabilities, and that it enables increased use of smart contracts for more complex algorithms, thus increasing the overall richness of the Ethereum ecosystem.
We propose a collaborative 3D modeling system, which is based on the blockchain technology. Our approach uses the blockchain to communicate with modeling tools and to provide them a decentralized database of the mesh modification history. This approach also provides a server-less version control system: users can commit their modifications to the blockchain and checkout others' modifications from the blockchain. As a result, our system enables users to do collaborative modeling without any central server.