Comparison of fixed-pattern and multiple-pattern structured light imaging systems

Abstract

We present a quantitative comparison of a fixed-pattern structured light system and a multi-pattern structured light system under varying capture environments. Several factors affect the performance of these systems, which makes the task of a fair comparison a very challenging aspect of this study. We conducted our experiments under controlled environment to enable us to control various system parameters for a fair comparison of these two techniques. We describe our methodology in choosing the system parameters for our study. For this analysis, we used two ground truth models with various depth and spatial variations as well as some smooth regions. These models are representative of two extremes of depth measurement scenarios. We show that multi-pattern approaches can be very accurate in controlled environment in stationary scenes due to high SNR, whereas fixed pattern methods are robust to ambient lighting changes but they have lesser accuracy. Further, in practical applications, we show that the multi-pattern approach has higher spatial and depth resolution when compared to a fixed pattern system.