New Software Converts 360 Videos Into 3D Models
Events
Subscribe:  iCal  |  Google Calendar
Birmingham GB   20, Sep — 24, Sep
Chiba JP   20, Sep — 24, Sep
Ohio US   21, Sep — 24, Sep
Los Angeles US   27, Sep — 1, Oct
Latest comments

Very interesting article, thanks for sharing!

astounding work there - hope someone important notices! I'm just jrpg fan. . . somebody needs to hire you!

by Sir Charles
8 hours ago

Yasss you guys rule so much LOVE please do the next part!!!

New Software Converts 360 Videos Into 3D Models
19 November, 2017
News

Maxime Lhuillier from CNRS/Institut Pascal/UCA demonstrated a technology that is capable of creating a 3D model of a space from a 360 video taken by a simple camera. Sounds crazy, right?

The process is very simple. Maxime Lhuillier used a Garmin Virb 360 camera to record a 360 video.  Then, the software generated a 3D model of a space as complex as a forest using that recording. Yes, the image quality is poor compared to results given by professional cameras. What is more, backlit portions of the videos have less visible detail after image compression. Still, just think about converting a 360 video into a 3D model with a simple laptop!

There are several limitations, though. A scene should be static with sufficient texture and light. Also, the camera motion must be slow enough and you should walk slowly. The idea here is not to get a perfect 3D scene, but to visualize a space using consumer cameras.

Check out these two models generated from videos taken by a 360 camera rig consisting of four helmet-mounted Gopro Hero 3 cameras recording video at 100 fps with frame-accurate synchronization:

The software and method are still in the works and the team is working hard to improve the results. You cannot try the software for yourself right now, so stay tuned for more details!

Comments

Leave a Reply

1 Comment on "New Software Converts 360 Videos Into 3D Models"

avatar
MeFeu
Guest
MeFeu

That’s basically what you can already do with photogrammetry. Just export your video into one file per frames, make photogrammetry out of it, and voilà. Nothing groundbreaking imo.

wpDiscuz