This dataset was created as part of research at Technion - Israel institute of technology

Note: Depth PNG images may look completely black when opened with a simple image viewing program due to the 16bit encoding.

Video 1

30 frames

Motion type: turn, parallel

Video 2

21 frames

Motion type: parallel

Video 3

30 frames

Motion type: turn

Video 4

20 frames

Motion type: parallel

Video 5

11 frames

Motion type: turn, parallel, perpendicular


  • Real sensor input from Kinect 1, Kinect 2 and RealSense R200 sensors.

  • RGB and depth images registered pixel-to-pixel versus high quality ground truth depth.

  • For each sensor there are two viewpoints:

    • RGB sensor with the registered depth image. This allows testing upsampling methods when the RGB image is substantially larger.

    • IR sensor with the registered RGB image for straightforward depth restoration.

  • The dataset consists five multiple frame videos, with varying types of object motion.


DROT is a depth dataset created to test depth restoration, rectification and upsampling methods.



I'd be glad to receive questions or suggestions you have about this dataset. Please feel free to leave your feedback or message me directly to the email below

dannecrot <at> gmail <dot> com

©2017 by Depth Restoration Occlusionless Temporal Dataset. Proudly created with

This site was designed with the
website builder. Create your website today.
Start Now