KinectFusion HQ

We present KinectFusion, a system that takes live depth data from a moving depth camera and in real-time creates high-quality 3D models. The system allows the user to scan a whole room and its contents within seconds. As the space is explored, new views of the scene and objects are revealed and these are fused into a single 3D model. The system continually tracks the 6DOF pose of the camera and rapidly builds a volumetric representation of arbitrary scenes.
Our technique for tracking is directly suited to the point-based depth data of Kinect, and requires no feature extraction or feature tracking. Once the 3D pose of the camera is known, each depth measurement from the sensor can be integrated into a volumetric representation. We describe the benefits of this representation over mesh-based approaches. In particular, the representation implicitly encodes predictions of the geometry of surfaces within a scene, which can be extracted readily from the volume. As the camera moves through the scene, new depth data can be added or removed from this volumetric representation, continually refining the 3D model acquired. We describe novel GPU-based implementations for both camera tracking and surface reconstruction. These take two well-understood methods from the computer vision and graphics literature as a starting point, defining new instantiations designed specifically for parallelizable GPGPU hardware. This allows for interactive real-time rates that have not previously been demonstrated.
We demonstrate the interactive possibilities enabled when high-quality 3D models can be acquired in real-time, including: extending multi-touch interactions to arbitrary surfaces; advanced features for augmented reality; real-time physics simulations of the dynamic model; novel methods for segmentation and tracking of scanned objects

Category:

Science & Technology

Tags:

• Microsoft
• Research
• MSR
• Kinect

[liual90]

omg please take my money already

[Haroids]

This video was rendered with a potato.

[photogomes]

@abram730 i love to construct cheeseburgers. I've worked at McDonald's when i was a teenager while i was delivering the community newspaper.

[abram730]

@photogomes "So my 14 years of modeling skills is obsolete"

Would you ever consider cheeseburger construction for change in careers?

[abram730]

@Parson2 Nope, skynet would be to obvious.. Making dues ex real, now that will slip under the radar.

watch?v=1b-bijO3uEw

[goomee3d]

now if they would put an HD Camera on the damned thing

[killerbee230]

Wow this is really really impressive! Looking forward to see more, good job guys/gals, would love to be part in such a project :)

[ChuckVanDamme]

This is by far the most amazing thing. Go, Microsoft!

[imbusy111]

"requires no feature extraction or feature tracking" - forgetting the fact that Kinect itself is doing feature tracking?

[theIrishFooFighter]

:O thats feckin unreal!!!

[kanabalize]

Hi may i know what programming language you used?

[philnolan3d]

@RelinquishLove I know it was intended for that purpose, but I really think this has more potential for 3D artists than anything else, therefore Sony is not really involved unless you mean for the artists that work at the Sony owned studios.