ABSTRACT
Indoor environment is the main area of people’s work and life. Indoor modeling is to establish the vectorized description of the walls, ceiling, floor, windows, doors, and other building structures or furniture features. This chapter presents the recent advances in indoor modeling with laser scanning techniques. This chapter focuses on two essential parts: calibration for multibeam laser scanners and indoor building modeling using a backpacked laser scanning point cloud. We first propose a target-free automatic self-calibration approach for multibeam laser scanners. The proposed method uses the isomorphism constraint among laser data to optimize the calibration parameters and then uses the ambiguity judgment algorithm to solve the mismatch problem. We then present a semantic line framework-based modeling building method for laser scanning point cloud data. The proposed method first semantically labels the raw point clouds into the walls, ceiling, floor, and other objects. Then line structures are extracted from the labeled points to achieve an initial description of the building line framework. To optimize the detected line structures caused by occlusion, a conditional generative adversarial nets (cGAN) deep learning model is constructed. The line framework optimization model includes structure completion, extrusion removal, and regularization. The result of the optimization is also derived from a quality evaluation of the point cloud. Thus, the data collection and building model representation become a united task-driven loop. The proposed method eventually outputs a semantic line framework model and provides a layout for the interior of the building.
