Active Image-based Modeling
share
We seek to automate the data capturing process in image-based modeling, which is often tedious and time consuming now. At the heart of our system is an iterative linear method to solve the multi-view stereo (MVS) problem quickly. Unlike conventional MVS algorithms that solve a per-pixel depth at each input image, we represent the depth map at each image as a piecewise planar triangle mesh and solve it by an iterative linear method. The edges of the triangle mesh are snapped to image edges to better capture scene structures. Our fast MVS algorithm enables online model reconstruction and quality assessment to determine the next-best-views (NBVs) for modeling. The NBVs are searched in a plane above unreconstructed shapes. In this way, our path planning can use the result from 3D reconstruction to guarantee obstacle avoidance. We test this system with an unmanned aerial vehicle (UAV) in a simulator, an indoor motion capture system (Vicon) room, and outdoor open spaces.
READ FULL TEXT