In the field of image synthesis, entanglement is the enmeshing of data properties with data other properties, which can make it difficult or impossible to isolate a particular aspect. If you change that data, you end up changing ‘nearby’ data. It’s not a trivial challenge to solve.
If time were no object, Neural Radiance Fields (NeRF) might by now have made greater inroads into potential commercial implementations – particularly in the field of human avatars and facial recreation.
New research from the University of California at Berkeley improves notably on recent efforts to create AI-powered image-editing procedures – this time by combining the considerable calculative forces of OpenAI’s GPT-3 Natural Language Processing (NLP) model with the latest version of Stability.ai’s world-storming Stable Diffusion text-to-image and image-to-image latent diffusion architecture.
In the 2014 Christmas special of the dystopian sci-fi anthology series Black Mirror, writer and series creator Charlie Brooker envisaged the possibility of cybernetic augmentations capable of ‘blocking out’ upsetting or banned material for a particular user – not only online, but in real-world interactions. In the episode, one of the main characters was ‘blocked’ by his ex-wife, so that she could literally no longer ‘see’ him.
The latent space is the ‘subconscious’ and overarching understanding of relationships between learned data points that a machine learning system has been able to derive from the information that it gets fed. This article takes a detailed look at what can be achieved by targeting content that’s been trained into the latent space of a machine learning model.
A new collaboration between Binghamton University and Intel offers a novel take on the problem of the unauthorized use of social media photos for facial recognition purposes, as well as AI and deepfake training – by using deepfake techniques to subtly alter the appearance of people in posted photos, so that only their friends and permitted contacts are able to see their original face images.
Besides their capacity to rip off the style of popular real world artists, the new breed of latent diffusion- based image synthesis systems promises a revolutionary ease-of-use not only for valid creative purposes, such as concept art development and stock image generation, but also for creating controversial deepfakes, and potentially objectionable imagery, including child pornography.
Over the past 10-20 years, and particularly in recent years, the computer vision research community has produced an abundance of frameworks capable of taking a single image and using it to perform ‘deepfake puppetry’ – the use of the facial and body movements of one person to simulate a secondary, fictitious identity.
New research from Tel Aviv University may prove capable of bringing authentic human motion to text-to-video synthesis, videogames, motion capture architectures in VFX pipelines, and also function as a synthetic data generator for downstream research initiatives, among a myriad of other potential applications – all thanks to what has turned out to be the most sensational AI technology of 2022 – diffusion models.
It’s nearly five years since the advent of deepfakes released into the public realm the ability to alter people’s facial identities; at first, in recorded video, and now even as a streaming implementation, with DeepFaceLive.
