Awesome Digital Imaging

This impressive yet creepy AI tech can take a single still image and make it appear to speak or sing any audio. Developed by researchers from Alibaba’s Institute for Intelligent Computing, EMO: Emote Portrait Alive’s model was trained on more than 150 million images to work its magic. The video is peppered with examples of the convincing deepfakery.

VFX artists have been making actors look younger or older for a while, but the effect is costly and labor-intensive. Researchers from Disney, ETH Zurich, and UW-Madison are showing off an AI tech called Face Re-aging Network (FRAN), which can perform the feat in five seconds per frame. It seems better at aging than de-aging, though.

CGI technology keeps getting more impressive. But it still takes a lot of time and effort to create photorealistic environments. Wren from Corridor Crew introduces us to a tech called Neural Radiance Fields, aka NeRFs, which can produce a texture-mapped 3D environment with realistic lighting from 2D photographs.

Two Minute Papers shows off a fascinating technology that can identify patterns in images then move them in the direction it thinks are most appropriate for their shape. The research paper is available here, and you can play with a version of the tech in the cinemagraph app Motionleap.

There’s software out there that can stylize images, but this technology makes images from real wood. Its algorithms compare greyscale levels between images and patterns in wood veneer, then instructs a CNC machine to cut out pieces that can be assembled to form a complete portrait. Two Minute Papers explains.

If you’ve used Photoshop’s content-aware fill, you know that it’s gotten easier to remove objects from still images. Doing the same in video is much trickier, but as Two Minute Papers explains, there’s new AI-based tech that’s really good at removing objects from moving images. It can also expand content into missing areas.