The method and system disclosed herein presents a method and system for capturing, by a depth-perception device on a mobile robot moving in an environment, depth-perception information at a first location within a portion of the environment. The method includes capturing, by a camera device on the mobile robot, an image frame at the first location within the portion of the environment, generating a three-dimensional point cloud representing the portion of the environment based on the depth-perception information and the image frame, extracting semantic information by providing the frame and the point cloud to a neural network model configured to perform semantic segmentation, associating the semantic information to the three-dimensional point cloud, projecting the three-dimensional point cloud onto a two-dimensional plane to form a first two-dimensional map; and demarcating the first two-dimensional map with a plurality of different regions based on the semantic information.