InLUT3D: Indoor Lodz University of Technology Point Cloud Dataset

Background

This resource contains Indoor Lodz University of Technology Point Cloud Dataset (InLUT3D) - a point cloud dataset tailored for real object classification and both semantic and instance segmentation tasks. Comprising of 321 scans, some areas in the dataset are covered by multiple scans. All of them are captured using the Leica BLK360 scanner.

Train/test split

The datset's authors impose the following train-test split:

Split
Setup range

train
from setup_0 to setup_300

test
from setup_301 to setup_320

The corresponding setups' objects are recommended for splits for the classificcation task.

Available categories

The points are divided into 18 distinct categories outlined in the label.yaml file along with their respective codes and colors. Among categories you will find:

ceiling,

floor,

wall,

stairs,

column,

chair,

sofa,

table,

storage,

door,

window,

plant,

dish,

wallmounted,

device,

radiator,

lighting,

other.

Challenges

Several challenges are intrinsic to the presented dataset:

Extremely non-uniform categories distribution across the dataset.

Presence of virtual images, particularly in reflective surfaces, and data exterior to windows and doors.

Occurrence of missing data due to scanning shadows (certain areas were inaccessible to the scanner's laser beam).

High point density throughout the dataset.

Dataset structure

The structure of the dataset is the following:

inlut3d.tar.gz/├─ setup_0/│  ├─ projection.jpg│  ├─ segmentation.jpg│  ├─ setup_0.pts├─ setup_1/│  ├─ projection.jpg│  ├─ segmentation.jpg│  ├─ setup_1.pts...

projection.jpg
A file containing a spherical projection of a corresponding PTS file.

segmentation.jpg
A file with objects marked with unique colours.

setup_x.pts
A file with point cloud int the textual PTS format.

Point characteristic

Each PTS file contains 8 columns:

Column ID
Description

1
X Cartesian coordinate

2
Y Cartesian coordinate

3
Z Cartesian Coordinate

4
Red colour in RGB space in the range [0, 255]

5
Green colour in RGB space in the range [0, 255]

6
Blue colour in RGB space in the range [0, 255]

7
Category code

8
Instance ID