TOF Range Enhancement

An important field of reasearch in computer vision is the 3D analysis and reconstruction of objects and scenes. A rather new technologie in this context is the PMD, based on the time-of-flight principle, which measures full-range distance information in real-time. Unfortunately, PMD-based devices have still limited resolution and provide only IR intensity information.

This reasearch direction focuses on a fast algorithmic approach to combine high resolution RGB images with PMD distance data, acquired using a binocular camera setup. The resulting combined RGBZ-data not only enhances the visual result, but also represents a basis for advanced data processing in e.g. object recognition with sub-pixel accuracy. A simple but efficient method is used to detect geometric occlusion caused by the binocular setup, which otherwise will lead to false color assignments.

Dynamic environements could cause mismatching regarding fusion. As any optical camera, ToF cameras are also subject to motion blur on dynamic scenes also known as motion artifacts. Here a GPU optical flow based method is proposed in order to warp all sequential sub phase images in the same reference in order to correct motion artifacts. A earlier method consisted to compute 3 optical flows and was achieving good results with a low framerate (around 12FPS). By reducing the number of optical flow computed and with a better memory management, a rate greater than 25FPS can be achieved.

This research project is partially funded by grant KO-2960-5/1,2 from the German Research Foundation (DFG).