April 27, 2026
Netflix has been working on using AI in TV shows and movies for several years via its Eyeline Studios team. Last year, an example of Netflix’s AI plans was revealed in the series, The Eternaut, during a VFX scene showing the demolition of a building. Now the streaming company is unveiling more of its AI roadmap thanks to contributions the Eyeline team made to GitHub.
The AI tool is called Vista4D and allows the user to take video footage and easily recreate the footage using different camera angles, framing, and movements.
“Vista4D is a video reshooting framework which synthesizes the dynamic scene represented by an input source video from novel camera trajectories and viewpoints,” reads the description of the tool’s function. “We bridge the distribution shift between training and inference for point-cloud-grounded video reshooting, as Vista4D is robust to point cloud artifacts from imprecise 4D reconstruction of real-world videos by training on noisy, reconstructed multiview videos.”
This tool could dramatically lower costs on a wide range of Netflix productions. By eliminating the need for reshoots, nearly every scene could become a one-shot take and serve as source material for the Vista4D system. However, the creators of the AI model admit that it could also be used in less-than-positive ways.
“Vista4D inherits the ethical questions which come with large generative models,” reads the research paper. “Enabling camera control over any video can have a profound effect on the emotional impact and public perception of the video, raising ethical concerns about content ownership and transformative work despite the creative possibilities our method opens.”
What’s also interesting is how this will blur the lines between what is and isn’t “AI video” since the end result wouldn’t be purely AI-generated video, but could rather be described as augmented real-world video. In recent months, the public and some Hollywood insiders have expressed negative opinions regarding AI video. But if the source video is real and simply augmented or manipulated by AI, that could muddy the water enough that debates around AI video could become hard to nail down, much less be noticed or critiqued by the general viewing public.
There’s no word on when this tool will be implemented in Netflix film and TV productions, or if it already has, but Eyeline will likely show off what it has accomplished when the time is right.
Cover image via Eyeline/Netflix
