Image-based panoramic 3D virtual environment using rotating two multi-view cameras

In this paper, we propose a new method for generating an image-based 3 0 panoramic virtual environment (VE). The panoramic VE is generated using 3 0 depth informa-tion estimated from rotating two multi-siew cameras. Even though conventional 211 image-based mosaicking niethods provide a wide view, they have limitations in providing a user with a navigation-enabled virtual en\+ ronment. In order to resolve such obstacles, weprst esti-mate the depth of the scene using two calibrated multi-view cameras and then stitch 31) point clouds instead of images. B v rolating hvo cameras using a turn-table it enables users to navigate the resulting 3 0 virtual envi-ronment with HMD. 1