Hugging Face's logo

albrateanu
/
ModalFormer

Image-to-Image
Transformers
low-light
low-light-image-enhancement
image-enhancement
image-restoration
computer-vision
low-light-enhance
multimodal
multimodal-learning
transformer
vision-transformer
vision-transformers
Eval Results
Model card Files
xet
Community
1
albrateanu commited on
Commit
e8627e3
·
verified ·
1 Parent(s): dea3ad3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +111 -110
README.md CHANGED
@@ -1,116 +1,117 @@
1
  ---
2
  license: mit
3
  tags:
4
- - low-light
5
- - low-light-image-enhancement
6
- - image-enhancement
7
- - image-restoration
8
- - computer-vision
9
- - low-light-enhance
10
- - multimodal
11
- - multimodal-learning
12
- - transformer
13
- - transformers
14
- - vision-transformer
 
15
  model-index:
16
- - name: ModalFormer
17
- results:
18
- - task:
19
- type: low-light-image-enhancement
20
- dataset:
21
- name: LOL-v1
22
- type: LOL-v1
23
- metrics:
24
- - name: PSNR
25
- type: PSNR
26
- value: 27.97
27
- - name: SSIM
28
- type: SSIM
29
- value: 0.897
30
- - task:
31
- type: low-light-image-enhancement
32
- dataset:
33
- name: LOL-v2-Real
34
- type: LOL-v2-Real
35
- metrics:
36
- - name: PSNR
37
- type: PSNR
38
- value: 29.33
39
- - name: SSIM
40
- type: SSIM
41
- value: 0.915
42
- - task:
43
- type: low-light-image-enhancement
44
- dataset:
45
- name: LOL-v2-Synthetic
46
- type: LOL-v2-Synthetic
47
- metrics:
48
- - name: PSNR
49
- type: PSNR
50
- value: 30.15
51
- - name: SSIM
52
- type: SSIM
53
- value: 0.951
54
- - task:
55
- type: low-light-image-enhancement
56
- dataset:
57
- name: SDSD-indoor
58
- type: SDSD-indoor
59
- metrics:
60
- - name: PSNR
61
- type: PSNR
62
- value: 31.37
63
- - name: SSIM
64
- type: SSIM
65
- value: 0.917
66
- - task:
67
- type: low-light-image-enhancement
68
- dataset:
69
- name: SDSD-outdoor
70
- type: SDSD-outdoor
71
- metrics:
72
- - name: PSNR
73
- type: PSNR
74
- value: 31.73
75
- - name: SSIM
76
- type: SSIM
77
- value: 0.904
78
- - task:
79
- type: low-light-image-enhancement
80
- dataset:
81
- name: MEF
82
- type: MEF
83
- metrics:
84
- - name: NIQE
85
- type: NIQE
86
- value: 3.44
87
- - task:
88
- type: low-light-image-enhancement
89
- dataset:
90
- name: LIME
91
- type: LIME
92
- metrics:
93
- - name: NIQE
94
- type: NIQE
95
- value: 3.82
96
- - task:
97
- type: low-light-image-enhancement
98
- dataset:
99
- name: DICM
100
- type: DICM
101
- metrics:
102
- - name: NIQE
103
- type: NIQE
104
- value: 3.64
105
- - task:
106
- type: low-light-image-enhancement
107
- dataset:
108
- name: NPE
109
- type: NPE
110
- metrics:
111
- - name: NIQE
112
- type: NIQE
113
- value: 3.55
114
  ---
115
 
116
  # ✨ ModalFormer: Multimodal Transformer for Low-Light Image Enhancement
@@ -149,4 +150,4 @@ Please check out the [**GitHub repository**](https://github.com/albrateanu/Modal
149
  ## 🙏 Acknowledgements
150
  We use [this codebase](https://github.com/caiyuanhao1998/Retinexformer) as foundation for our implementation.
151
 
152
- Paper: https://arxiv.org/pdf/2507.20388
 
1
  ---
2
  license: mit
3
  tags:
4
+ - low-light
5
+ - low-light-image-enhancement
6
+ - image-enhancement
7
+ - image-restoration
8
+ - computer-vision
9
+ - low-light-enhance
10
+ - multimodal
11
+ - multimodal-learning
12
+ - transformer
13
+ - transformers
14
+ - vision-transformer
15
+ - vision-transformers
16
  model-index:
17
+ - name: ModalFormer
18
+ results:
19
+ - task:
20
+ type: low-light-image-enhancement
21
+ dataset:
22
+ name: LOL-v1
23
+ type: LOL-v1
24
+ metrics:
25
+ - name: PSNR
26
+ type: PSNR
27
+ value: 27.97
28
+ - name: SSIM
29
+ type: SSIM
30
+ value: 0.897
31
+ - task:
32
+ type: low-light-image-enhancement
33
+ dataset:
34
+ name: LOL-v2-Real
35
+ type: LOL-v2-Real
36
+ metrics:
37
+ - name: PSNR
38
+ type: PSNR
39
+ value: 29.33
40
+ - name: SSIM
41
+ type: SSIM
42
+ value: 0.915
43
+ - task:
44
+ type: low-light-image-enhancement
45
+ dataset:
46
+ name: LOL-v2-Synthetic
47
+ type: LOL-v2-Synthetic
48
+ metrics:
49
+ - name: PSNR
50
+ type: PSNR
51
+ value: 30.15
52
+ - name: SSIM
53
+ type: SSIM
54
+ value: 0.951
55
+ - task:
56
+ type: low-light-image-enhancement
57
+ dataset:
58
+ name: SDSD-indoor
59
+ type: SDSD-indoor
60
+ metrics:
61
+ - name: PSNR
62
+ type: PSNR
63
+ value: 31.37
64
+ - name: SSIM
65
+ type: SSIM
66
+ value: 0.917
67
+ - task:
68
+ type: low-light-image-enhancement
69
+ dataset:
70
+ name: SDSD-outdoor
71
+ type: SDSD-outdoor
72
+ metrics:
73
+ - name: PSNR
74
+ type: PSNR
75
+ value: 31.73
76
+ - name: SSIM
77
+ type: SSIM
78
+ value: 0.904
79
+ - task:
80
+ type: low-light-image-enhancement
81
+ dataset:
82
+ name: MEF
83
+ type: MEF
84
+ metrics:
85
+ - name: NIQE
86
+ type: NIQE
87
+ value: 3.44
88
+ - task:
89
+ type: low-light-image-enhancement
90
+ dataset:
91
+ name: LIME
92
+ type: LIME
93
+ metrics:
94
+ - name: NIQE
95
+ type: NIQE
96
+ value: 3.82
97
+ - task:
98
+ type: low-light-image-enhancement
99
+ dataset:
100
+ name: DICM
101
+ type: DICM
102
+ metrics:
103
+ - name: NIQE
104
+ type: NIQE
105
+ value: 3.64
106
+ - task:
107
+ type: low-light-image-enhancement
108
+ dataset:
109
+ name: NPE
110
+ type: NPE
111
+ metrics:
112
+ - name: NIQE
113
+ type: NIQE
114
+ value: 3.55
115
  ---
116
 
117
  # ✨ ModalFormer: Multimodal Transformer for Low-Light Image Enhancement
 
150
  ## 🙏 Acknowledgements
151
  We use [this codebase](https://github.com/caiyuanhao1998/Retinexformer) as foundation for our implementation.
152
 
153
+ Paper: https://arxiv.org/pdf/2507.20388