Robust Plane Detection Using Depth Information From a Consumer Depth Camera

摘要: The emerging of depth-camera technology is paving the way for a variety new applications and it believed that plane detection one them. In fact, planes are common in man-made living structures, thus their accurate can benefit many visual-based applications. use depth information allows detecting characterized by complex pattern texture, where texture-based algorithms usually fail. this paper, we propose robust depth-driven (DPD) algorithm which consists two parts: growing-based two-stage refinement. proposed approach starts from seed patch with highest planarity uses estimated equation growing dynamic threshold function to steer process. Aided mechanism, each grow its maximum extent, then next grow. This process iteratively repeated so as detect all planes. Moreover, refinement tackle problems suffered approaches, over-growing problem, under-growing problem. Validated extensive experiments, DPD able accurately various testing conditions. terms applications, be used pre-processing step such as, planar object recognition, super-resolution time-of-flight images intrinsically low resolution.