Hugging Face's logo

Datasets:
tanvirnwu
/
HazeSpace2M

Tasks:
Image Classification
Modalities:
Image
Sub-tasks:
multi-class-image-classification
Languages:
English
Size:
100B<n<1T
Tags:
image
dehazing
dehazer
classification
haze
hazy
License:
Dataset card Files Community
HazeSpace2M
File size: 2,834 Bytes 
5ac94b0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b1ed3c8
 
5ac94b0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
 
---
license: cc-by-4.0
language:
- en
tags:
- image
- dehazing
- dehazer
- classification
- haze
- hazy
- hazespace2m
- removal
- enhancement
- restoration
- image restoration
- image enhancement
- single image dehazing
- multi weather dehazing
- dehazing dataset

pretty_name: >-
  Single Image Dehazing Dataset with Over 2 Million Hazy Images of three
  different types of hazes such as fog, cloud, and environmental haze (EH) of 10
  different intense levels.
size_categories:
- 100B<n<1T

task_categories:  # Full list at https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/src/pipelines.ts
- image-classification
task_ids:
- multi-class-image-classification
---


<h4 align="center"><strong><a href="https://2024.acmmm.org/">Published in ACM Multimedia 2024, Melbourne, Australia</a></strong></h4>
<h2 align="center"><strong>HazeSpace2M: A Dataset for Haze Aware Single Image Dehazing <a href="https://tanvirnwu.github.io/assets/papers/HazeSpace2M.pdf" target="_blank">[Paper]</a></strong></h2>
<h6 align="center">Md Tanvir Islam<sup> 1</sup>, Nasir Rahim<sup> 1</sup>, Saeed Anwar<sup> 2</sup>, Muhammad Saqib<sup> 3</sup>, Sambit Bakshi<sup> 4</sup>, Khan Muhammad<sup> 1, *</sup></h6>
<h6 align="center">| 1. Sungkyunkwan University, South Korea | 2. KFUPM, KSA | 3. UTS, Australia | 4. NIT Rourkela, India || *Corresponding Author |</h6> 
<hr>

# IMPORTANT UPDATES
- 2025/03/22 | Fixed the indexing issues of the [Farmland](https://huggingface.co/datasets/tanvirnwu/HazeSpace2M/tree/main/Farmland) subset.
- 2025/03/21 | Identified an indexing issue between GT and Haze images. We are working on it to fix the issues.
<hr>

<h1><b></b>Dataset Description</b></h1>
<b>GitHub Repository:</b> https://github.com/tanvirnwu/HazeSpace2M

<b>Paper:</b> https://dl.acm.org/doi/abs/10.1145/3664647.3681382

<b>Point of Contact:</b> [email protected]

<b>Dataset Size:</b> Outdoor: 269GB | Street: 295GB | Farmland: 90GB | Satellite: 153GB

<hr>

# HazeSpace2M Dataset
![](https://raw.githubusercontent.com/tanvirnwu/HazeSpace2M/master/assets/HazeSpace2M.jpg)

<hr>

# HazeSpace2M Folder Structure
![](https://raw.githubusercontent.com/tanvirnwu/HazeSpace2M/master/assets/datasetStructure-02.jpg)

<hr>

# Proposed Multi-stage Haze Aware Dehazing
![](https://raw.githubusercontent.com/tanvirnwu/HazeSpace2M/master/assets/proposedFramework.jpg)

<hr>

# Cite This Paper

If you find our work useful in your research, please consider citing our paper:

```bibtex
@inproceedings{hazespace2m,
  title={HazeSpace2M: A Dataset for Haze Aware Single Image Dehazing},
  author={Islam, Md Tanvir and Rahim, Nasir and Anwar, Saeed and Saqib Muhammad and Bakshi, Sambit and Muhammad, Khan},
  booktitle={Proceedings of the 32nd ACM International Conference on Multimedia},
  year={2024},
  doi = {10.1145/3664647.3681382}
}