Data-Fusion of PMD-Based Distance-Information and High-Resolution RGB-Images

Abstract

An important field of reasearch in computer vision is the 3D analysis and reconstruction of objects and scenes for e.g. position determination, online object recognition or collision prevention. Known techniques thereby mainly rely on measuring devices such as laser scanners, stereo camera systems, or comparable algorithms like structure from motion. The rather new Photonic Mixing Device (PMD) technique is based on the time-of-flight principle and measures full-range distance information in real-time. Unfortunately, PMD-based devices have still limited resolution (e.g. 160 x 120px) and provide only graylevel information. This paper describes a fast algorithmic approach to combine high resolution RGB images with low resolution PMD distance data, acquired using a binocular camera setup. The approach relies completely on fast parallel Graphics Processing Units (GPUs) ressources. Additionally, we introduce a simple but efficient method to detect insufficient color assignments, which are due to the binocular 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.