AI tools are currently dominating the scene,with companies like Google and OpenAI pushing the boundaries with creations like Sora, which can not only generate images and videos but also produce highly shareable deepfakes, leading to recent controversies involving Microsoft.
A captivating video has recently gone viral, showcasing a clever transformation where rapper Lil Yachty's live performance is seamlessly replaced with Joaquin Phoenix's iconic portrayal of The Joker. AI enthusiast Barsee shared the video on social media, revealing that the deepfake was crafted using an AI tool capable of replacing individuals within existing video footage.
However, the true intricacy behind the video's production extends beyond the mere utilization of a single AI tool, hinting at a more sophisticated and multi-step process. The creation of such a deepfake involves a combination of techniques and tools, requiring a deep understanding of AI algorithms and video editing skills. This highlights the growing complexity and potential of AI-driven deepfake technology.
The captivating deepfake video, which replaces rapper Lil Yachty's live performance with Joaquin Phoenix's portrayal of The Joker, was masterfully crafted by a user named A.I Warper. To achieve this stunning result, A.I Warper employed a combination of tools, including Viggle AI, several other AI models, and Adobe After Effects.
Viggle AI, as described on its official website, harnesses "the first video-3D foundation model with actual physics understanding." A.I Warper's methodology involved rotoscoping Lil Yachty to create a mask, which was then used as a guide. Next, an image of The Joker was fed into Viggle AI to seamlessly replace Lil Yachty with the iconic character. The resulting output was further refined by applying an "AnimateDiff" motion model, alongside other AI tools specifically trained on Joaquin Phoenix's Joker, to achieve the final, jaw-dropping transformation.
Although the video reportedly took less than five minutes to create, the complexity of the tools and expertise required to produce such a deepfake should not be underestimated. Viggle AI alone was not enough; the rotoscoped output had to be meticulously composited back into the original video and then further polished to achieve the impressive final result. Despite the brief creation time, the end product is undeniably awe-inspiring, showcasing the incredible potential of AI-driven deepfakes.