RGB-D Kitchen Dataset

Paired RGB and depth video from real kitchen environments with registered depth maps and 3D annotations for training depth-aware kitchen robots.

Dataset at a Glance

50K+
Video clips
350+
Hours recorded
30+ kitchens
Environments
6+
Annotation layers

Comparison with Public Datasets

How Claru's dataset compares to publicly available alternatives.

DatasetClipsHoursModalitiesEnvironmentsAnnotations
ScanNet1.5K scans~5RGB-D707 rooms3D segmentation
NYU Depth V21.4K~2RGB-D464 scenesDepth, segmentation
Claru RGB-D Kitchen50K+350+RGB-D30+ kitchensDepth, 3D reconstruction, objects, surfaces

Use Cases

Depth-Aware Manipulation

Using depth for reach planning and collision avoidance in cluttered kitchens. Example models: UniDepth, DepthAnything, ICP grasping.

3D Scene Understanding

Building 3D kitchen representations for spatial reasoning. Example models: ScanNet++, Habitat, iGibson.

Transparent Object Detection

Using depth discontinuities to detect glass and clear objects. Example models: ClearGrasp, TransCG, Dex-NeRF.

Key References

  1. [1]Dai et al.. ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes.” CVPR 2017, 2017. Link
  2. [2]Yang et al.. Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data.” CVPR 2024, 2024. Link
  3. [3]Sajjan et al.. ClearGrasp: 3D Shape Estimation of Transparent Objects.” ICRA 2020, 2020. Link

How Claru Delivers This Data

Claru captures synchronized RGB-D data using Intel RealSense cameras with factory-calibrated registration. Kitchen-specific depth data captures challenging geometry: reflective steel, transparent glass, and steam.

Frequently Asked Questions

Intel RealSense D435/D455 at 848x480 depth, synchronized with 1920x1080 RGB at 30fps.

Processing flags low-confidence depth regions and provides confidence maps. Transparent objects get supplementary boundary annotations.

Yes. Camera parameters enable point cloud generation and TSDF reconstruction. Pre-computed meshes available for a subset.

Request a Sample Pack

Get a curated sample of rgb-d kitchen data with full annotations to evaluate for your project.