A Real World Dataset for Multi-view 3D Reconstruction
This dataset addresses a gap for researchers in 3D computer vision by offering a new benchmark, though it is incremental as it builds on existing data collection efforts.
The authors tackled the lack of a real-world benchmark for learned multi-view 3D reconstruction by creating a dataset of 998 3D models with 847,000 annotated RGB and depth images, providing camera and object poses to facilitate applications like shape reconstruction and object pose estimation.
We present a dataset of 998 3D models of everyday tabletop objects along with their 847,000 real world RGB and depth images. Accurate annotations of camera poses and object poses for each image are performed in a semi-automated fashion to facilitate the use of the dataset for myriad 3D applications like shape reconstruction, object pose estimation, shape retrieval etc. We primarily focus on learned multi-view 3D reconstruction due to the lack of appropriate real world benchmark for the task and demonstrate that our dataset can fill that gap. The entire annotated dataset along with the source code for the annotation tools and evaluation baselines is available at http://www.ocrtoc.org/3d-reconstruction.html.