Semantic Scene Completion from a Single 360-Degree Image and Depth Map

摘要

We present a method for Semantic Scene Completion (SSC) of complete indoor scenes from a single 360◦ RGB image and corresponding depth map using a Deep Convolution Neural Network that takes advantage of existing datasets of synthetic and real RGB-D images for training. Recent works on SSC only perform occupancy prediction of small regions of the room covered by the field-of-view of the sensor in use, which implies the need of multiple images to cover the whole scene, being an inappropriate method for dynamic scenes. Our approach uses only a single 360º image with its corresponding depth map to infer the occupancy and semantic labels of the whole room. Using one single image is important to allow predictions with no previous knowledge of the scene and enable extension to dynamic scene applications. We evaluated our method on two 360º image datasets: a high-quality 360º RGB-D dataset gathered with a Matterport sensor and low-quality 360º RGB-D images generated with a pair of commercial 360º cameras and stereo matching. The experiments showed that the proposed pipeline performs SSC not only with Matterport cameras but also with more affordable 360º cameras, which adds a great number of potential applications, including immersive spatial audio reproduction, augmented reality, assistive computing and robotics.