Embedded Files
Seunghui Shin, Jaeyun Jang, Sundong Park, and Hyoseok Hwang*
{jumin1116, yoon2926, sundong, hyoseok}@khu.ac.kr
IEEE Robotics and Automation Letters (RA-L) 2025
Dataset Structure
Dataset Structure
This dataset provides Interference Affected Depth Data, Ground Truth, and additional data from Multiple Intel Realsense L515 Systems.
Data
Data
Raw Data (640 x 480)
Interference Unaffected Depth
Interference Unaffected RGB
Interference Affected Depth
Interference Affected RGB
Ground Truth (640 x 480)
Interference Artifact Mask
Sensor Artifact Mask
GT Depth Data
Categories (Number of Objects)
Categories (Number of Objects)
We Collected a dataset from various objects classified into four different categories.
Grocery (13 objects)
Kitchen (15 objects)
Office (18 objects)
Toy (17 objects)
Data Directory
Data Directory
Our dataset includes two kinds of data. Raw Dataset is using for training directly. Our model was trained with this format. Category Dataset is a dataset that categorizes scenes by a specified category. Users can select useful category and can customize it for training. The Category Dataset includes each camera's intrinsic parameters and extrinsic parameters from each scenes.
Both Raw Dataset and Category Dataset include several Challenging Scenarios. Our dataset directory structure is below.
By default, two color images are provided for each camera in the object directory of Category Dataset, one per "interference_unaffected" and one per "interference_affected", and an additional DRIM_RGB.zip file is provided that matches the number of color images to the number of depth data. We recommend downloading one of the two zip files provided to suit your needs.
Interference Affected Depth Data
Interference Affected Depth Data
License
License
The DRIM dataset is provided under the Open Database License (available here).
This means that you are free to use, distribute, modify and create derivative works from this dataset, provided that you give appropriate credit to our work.
In addition, any adapted version of this dataset that is used publicly must be released under the same licence, and any redistribution must remain open.
When using the DRIM dataset, please ensure that you acknowledge our work by citing the referenced paper.
Citation
Citation
Please use the following BibTeX entry to cite this work:
@article{shin2025drim,
title={DRIM: Depth Restoration with Interference Mitigation in Multiple LiDAR Depth Cameras},
author={Shin, Seunghui and Jang, Jaeyun and Park, Sundong and Hwang, Hyoseok},
journal={IEEE Robotics and Automation Letters},
year={2025},
publisher={IEEE}
}
Page updated
Google Sites
Report abuse