The Microsoft Kinect camera has caught the imagination of numerous developers who are using it to capture images of scenes that they can then reconstruct in 3-D.
However, structured-light based cameras like the Kinect have a limited field of view. The result is that when the view of a scene is synthesized from a new viewpoint, depth information is often missing.
Now, Microsoft researchers, working with engineers at the University of Science and Technology of China and Texas A&M University (College Station, TX, USA) have developed new software that has enabled them to reconstruct a 3-D scene from multiple structured-light based cameras.
The novel 3-D depth reconstruction algorithm fuses the results from multiple cameras, while combating the interference caused by multiple active sensors.
The researchers say that the solution can be applied to commodity depth cameras without special customization.
A detailed technical description of the research work can be found here.
Interested in reading how others are using the Kinect? Here’s a compendium of news stories on the Kinect that Vision Systems Design has published over the past year.
1. 3-D algorithms combine with Kinect to track objects using continuous point cloud data
Ryohei Ueda from the JSK laboratory at the University of Tokyo (Tokyo, Japan) has developed 3-D algorithms that can be used with the Microsoft Kinect to help computer systems track objects in a scene in real time.
2. Satellites dock with Microsoft's Kinect
Engineers at the University of Surrey (Guildford, UK) and Surrey Satellite Technology Limited (SSTL; Guildford, UK) are developing an in-orbit docking system for small satellites that is based upon Microsoft’s Kinect system.
3. Microsoft’s Kinect helps keep surgery sterile
Engineers based at Microsoft Research (Cambridge, UK) have developed a system based on the Microsoft Kinect that allows surgeons to manipulate images in the operating theater.
4. Shoppers create 3-D models of their own bodies online
A startup company established by a team that includes several alumni from the University of Cambridge’s Department of Engineering (Cambridge, UK) has developed technology intended to help shoppers buy clothes online.
5. MIT robot navigates using Microsoft’s Kinect
Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL; Cambridge, MA, USA) have developed a robot that uses Microsoft’s Kinect to navigate through its surroundings.
6. Kinect transformed into 3-D image capture system
Engineers at the University of California San Diego (UCSD) (San Diego, CA, USA) have modified Microsoft's Kinect Xbox 360 peripheral to enable it to make handheld 3-D scans of objects.
7. EVT combines EyeScan 3D system with Kinect sensor for industrial assembly and palletization
EVT (Karlsruhe, Germany) has expanded its EyeScan 3D smart camera system with a Microsoft (Redmond, WA, USA) Kinect sensor. This enables resolution of three-dimensional tasks in robotics applications, among them palletization and depalletization but also in the sorting of components on the assembly line.
8. MVTec tests Halcon machine-vision software with Microsoft Kinect for 3-D vision
Microsoft’s (Redmond, WA, USA) Kinect camera/sensor is being used by machine-vision technology supplier MVTec Software (Munich, Germany) to develop new, low-cost options for 3-D vision in industrial automation and research applications.
9. Scanners use light projectors for fast 3-D image reconstruction
In many industrial applications deployed for inspection and quality control, it is necessary to capture three-dimensional images of the object in order to rapidly analyze any structural defects.
10. 3-D image processing relieves male shoppers from drudgery
A London, UK-based outfit called Bodymetrics has taken eight Kinect for Windows sensors and integrated them into a 3-D body-mapping system called the Bodymetrics Pod.
-- by Dave Wilson, Senior Editor, Vision Systems Design