8 datasets found

The NYU Hand Pose Dataset comprises 70,000 images captured with a depth sensor in VGA resolution accompanied by ground truth annotations of positions of hand joints.

The Object Clutter Indoor Dataset (OCID) is a real-world RGB-D dataset for object instance segmentation.

The Tabletop Objects Dataset (TOD) is a synthetic RGB-D dataset for zero-shot instance segmentation. It consists of 40K scenes with 25 object categories.

A synthetic dataset for dynamic novel view synthesis, consisting of 3,000 scenes of 60 frames each, at a frame rate of 24 FPS, with RGB-D data rendered from 16 virtual cameras...

A large-scale dataset of indoor 3D scenes with semantic annotations.

The SUN3D dataset is a benchmark for 3D matching tasks. It contains 15,000 3D models with RGB-D data.

The 7-scenes dataset is a benchmark for 3D matching tasks. It contains 7 different scenes with RGB-D data.

The Replica dataset provides real-world scenes which we use to evaluate and compare our approach with related work. The dataset consists of high quality 3D reconstructions of 5...