Time-of-flight cameras
A new type of cameras have arrived which builds a 3D map of the environment by measuring the time for light to leave the camera, reflect and return. By measuring the time this takes, a 3D map can be built of the environment. This measurement principle allows for video-rate images containing both 3D scene structure and reflected intensity.
Time-of-flight cameras are a new type of sensors, with a working principle similar to radars and ladars in that they use time to determine object distance. Multiple vendors are now supplying such cameras, most of them using a common sine-based modulation scheme and measurement principle. Prices can be as low as €600 for a 64x48 pixel camera. This makes the technology affordable and implementable in real-world scenarios.
The main benefits of time-of-flight cameras as opposed to other 3D measurement techniques is that they
- provide the 3D information in realtime,
- are small and light
- do not have problems with occlusions (which triangulation-based systems can have)
This makes them well suited for 3D imaging of moving objects and dynamic scenes. SINTEF has experience in using these cameras for autonomous robot navigation and control. The 3D information that these cameras provide makes it possible to develop robust algorithms for providing the robot with information about its surroundings.
Possible other applications include:
- Surveillance and proximity monitoring
- Man-machine interaction
- Object tracking
- Eye-hand coordination for robots
- Person counting
The current generation of time-of-flight cameras introduce artefacts into the measured data. These artefacts may reduce the data quality of the 3D measurements, which means that special care needs to be taken when analyzing data from these sensors. SINTEF has researched and published methods within this field, especially with regard to effects occuring in scenes with non-trivial scene geometry.
For further information or inquiries regarding time-of-flight cameras, please contact Jens T Thielemann
.
Related projects
Related publications
Pipeline landmark detection for autonomous robot navigation using time-of-flight imagery
Thielemann, J.T.; Breivik, G.M.; Berge, A.
Computer Vision and Pattern Recognition Workshops, 2008. CVPR Workshops 2008. IEEE Computer Society Conference on
Volume, Issue, 23-28 June 2008 Page(s): 1 - 7
Modelling and Compensating Measurement Errors Caused by Scattering in Time-Of-Flight Cameras
Kavli Tom, Kirkhus Trine, Thielemann Jens T, Borys Jagielski Proceedings of SPIE, the International Society for Optical Engineering, 7066 August 2008.
