RGB-D Object Dataset

The dataset contains 300 objects (aka "instances") in 51 categories. The dataset contains several different types of data and they have been split up here so you can download them depending on your needs. Except the evaluation set, all of them will extract to a folder called rgbd-dataset but the names of the extracted files are all different, so it is possible to extract everything to a common root directory without overwriting. The evaluation set unpacks to a separate folder called rgbd-dataset_eval.

Cropped RGB and depth images with object segmentation masks

Segmented object 3D point clouds

Full 640x480 RGB and depth images with object segmentation masks

Approximate ground truth pose labels

Evaluation Set (Subsampled turntable frames as used in our RGB-D object recognition experiments)

RGB-D Scenes Dataset v.2

The RGB-D Scenes Dataset v2 consists of 14 scenes containing furniture (chair, coffee table, sofa, table) and a subset of the objects in the RGB-D Object Dataset (bowls, caps, cereal boxes, coffee mugs, and soda cans). Each scene is a point cloud created by aligning a set of video frames using Patch Volumes Mapping*. These 3D reconstructions and ground truth object annotations are exactly those used in our ICRA 2014 paper (see README).

RGB-D Scenes Dataset v.2 - Scene point clouds, RGB-D video frames, and Trimble 3D Warehouse objects

RGB-D Scenes Dataset

This dataset contains 8 scenes annotated with objects that belong to the RGB-D Object Dataset. Each scene is a single video sequence consisting of multiple RGB-D frames.

RGB-D Scenes - individual video frames

RGB-D Scenes - 3D reconstruction point clouds from aligned video frames

Detection-based Object Labeling in 3D Scenes

Detection-based Object Labeling in 3D Scenes
Kevin Lai, Liefeng Bo, Xiaofeng Ren, and Dieter Fox. ICRA 2012, May 2012.

Software and Data for the technique from the above paper is now available.