Skip to main content
  1. Home
  2. Computing
  3. News

MIT researchers develop simulation to manipulate real-world objects in videos

Add as a preferred source on Google

Pokémon Go just passed the 100 million player mark — and it’s still growing. The game’s success can be largely credited to millennial nostalgia but also to the app’s clever use of augmented reality (AR), which overlays the Pokémon universe on the real world via the smartphone’s camera.

But, even though you can toss a virtual Poké Ball at a virtual Pikachu on the sidewalk, traditional augmented reality doesn’t let virtual objects interact with objects in the real world. Pikachu can’t electrify a puddle, bump into a signpost, or jump up on the curb.

Recommended Videos

A team of Massachusetts Institute of Technology researchers led by Abe Davis, a Ph.D student at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL), and Justin Chen, a post-doctoral associate in its Department of Civil and Environmental Engineering, might help smooth these distinctions between augmented reality and the real world by developing Interactive Dynamic Video (IDV), which they think can change industries from engineering to entertainment.

Two years ago, Davis and his team recorded silent videos of objects that vibrated when they were hit by sound. Using a processing algorithm, the researchers were able to recover music and speech from these tiny vibrations — effectively eavesdropping through a bag of potato chips. This discovery enticed Davis to explore vibrations further and inspired him to see if an algorithm could re-create an object’s movement by analyzing the object’s supersmall vibrations.

“There are some classic techniques used in engineering and physically based animation to simulate objects, but they depend on a full 3D model,” Davis told Digital Trends. “I got this crazy idea that you could do something similar with a video instead, and it ended up working surprisingly well.”

Davis and his team used a traditional camera to record various objects as they vibrated ever so slightly. They recorded a bush as it blew in the breeze and a guitar string as it was lightly touched. The algorithm then identified “vibrations modes” — how a particular object might sway this way or rock that way when disrupted by vibrations. From here, the researchers can use a simulation to virtually interact with the object.

So how could this apply to Pokémon Go?

Pokemon GO and Interactive Dynamic Video

“There is a broad range of applications for this work,” Davis said, “from monitoring the safety of bridges and buildings, to product testing, to special effects for movies, and ‘dynamic augmented reality,'” which can interact with and respond to virtual forces.

Davis admits, however, that it will still be a while before this technology ends up on mobile devices. Instead, it will likely be applied to movie special effects, as well as structural health monitoring to predict the safety of old bridges.

So, although we won’t soon see Pikachu virtually jump-start our car, we may see IDV help make our world a little more safer.

Dyllan Furness
Former Contributor
Dyllan Furness is a freelance writer from Florida. He covers strange science and emerging tech for Digital Trends, focusing…
Google brings its best AI music model Lyria 3.5 to the Gemini app
Developers get full API access through Google AI Studio.
Google-gemini-lyria-3.5

Google has added Lyria 3.5, its most advanced music generation model yet, to the Gemini app. Previously available through Google's AI filmmaking tool Flow, the model is now rolling out to all Gemini users, making it easier to generate polished songs, instrumentals, and soundtracks from simple text prompts or even photos.

Lyria 3.5 makes AI-generated music sound more natural

Read more
AMD’s new workstation packs 576GB of GPU memory and enough horsepower for trillion-parameter AI models
AMD just built a desktop workstation with more raw memory than most small data centers had a decade ago.
People, Person, Performer

AMD just built a desktop that really has no business sitting on an actual desk. At IFA 2026, the company unveiled the Threadripper Halo Station.

As the name probably already suggests, it is a liquid-cooled machine that can run AI models with more than a trillion parameters without touching the cloud even once. If you don't already know, those numbers are reserved for server shelves, not someone's personal desk.

Read more
Why doesn’t DDR5 memory always perform the way you expect?
Electronics, Hardware, Computer Hardware

This post is brought to you in paid partnership with ASUS.

DDR5 memory doesn't always perform the way you'd expect because so much of its behavior is shaped by the platform it's running on, not the memory kit alone. A motherboard is one of the few PC components that continues to influence your system long after you've finished building it. Even after you've chosen your processor, graphics card, and memory, it continues to shape which upgrades are possible, how well new hardware behaves, and whether your PC evolves gracefully or starts showing its age before the rest of the system does.

Read more