Facing the sparse LiDAR (Light Detecting and Ranging) data of New Orleans area in America

a new residential building reconstruction method based on the fusion of LiDAR and satellite imagery is proposed.The main contribution of this work is the automatic isolation of roof points and roof type recognition.Using LiDAR data boundary to identify the ROI area in satellite imagery

and using cue lines extracted from ROI areas

the building roofs are isolated.Then

based on the relationship of normal vectors

building types are recognized

and then buildings are reconstructed.Experiments show that our method can successfully reconstruct residential buildings given relatively sparse LiDAR samples

achieve a high reconstruction rate in a reasonably short time

which meet the requirement of virtual reality systems.