Joint 2D-3D-Semantic Data for Indoor Scene Understanding
This dataset addresses the need for comprehensive indoor scene data for researchers in computer vision and robotics, though it is incremental as it builds on existing data collection efforts.
The authors introduced a large-scale indoor scene dataset with over 6,000m2 and 70,000 RGB images, featuring registered 2D, 2.5D, and 3D modalities and instance-level annotations to support joint and cross-modal learning models.
We present a dataset of large-scale indoor spaces that provides a variety of mutually registered modalities from 2D, 2.5D and 3D domains, with instance-level semantic and geometric annotations. The dataset covers over 6,000m2 and contains over 70,000 RGB images, along with the corresponding depths, surface normals, semantic annotations, global XYZ images (all in forms of both regular and 360° equirectangular images) as well as camera information. It also includes registered raw and semantically annotated 3D meshes and point clouds. The dataset enables development of joint and cross-modal learning models and potentially unsupervised approaches utilizing the regularities present in large-scale indoor spaces. The dataset is available here: http://3Dsemantics.stanford.edu/