Kim2091 commited on
Commit
4acdd36
·
verified ·
1 Parent(s): 77c109f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +19 -0
README.md CHANGED
@@ -6,28 +6,47 @@ license: cc-by-nc-sa-4.0
6
  ## 2x-AnimeSharpV3
7
 
8
  **Scale:** 2
 
9
  **Architecture:** ESRGAN
 
10
  **Links:** [Github Release](<https://github.com/Kim2091/Kim2091-Models/releases/tag/2x-AnimeSharpV3>)
11
 
 
12
  **Author:** Kim2091
 
13
  **License:** CC BY-NC-SA 4.0
 
14
  **Purpose:** Anime
 
15
  **Subject:**
 
16
  **Input Type:** Images
 
17
  **Date:** 10-24-24
 
18
  **Size:**
 
19
  **I/O Channels:** 3(RGB)->3(RGB)
20
 
 
21
  **Dataset:** ModernAnimation1080_v3
 
22
  **Dataset Size:** 2-3k
 
23
  **OTF (on the fly augmentations):** No
 
24
  **Pretrained Model:** 4xESRGAN
 
25
  **Iterations:** 140k
 
26
  **Batch Size:** 8
 
27
  **GT Size:** 64-128
28
 
 
29
  **Description:** I gave up on having multiple architectures. This release is ESRGAN only, and provides superior quality compared to AnimeSharpV2 in nearly every scenario. It has most of the advantages of the old V2 Sharp models, while not having issues with depth of field.
30
 
 
31
  __Comparisons:__ <https://slow.pics/c/A2BRSa0U>
32
 
33
 
 
6
  ## 2x-AnimeSharpV3
7
 
8
  **Scale:** 2
9
+
10
  **Architecture:** ESRGAN
11
+
12
  **Links:** [Github Release](<https://github.com/Kim2091/Kim2091-Models/releases/tag/2x-AnimeSharpV3>)
13
 
14
+
15
  **Author:** Kim2091
16
+
17
  **License:** CC BY-NC-SA 4.0
18
+
19
  **Purpose:** Anime
20
+
21
  **Subject:**
22
+
23
  **Input Type:** Images
24
+
25
  **Date:** 10-24-24
26
+
27
  **Size:**
28
+
29
  **I/O Channels:** 3(RGB)->3(RGB)
30
 
31
+
32
  **Dataset:** ModernAnimation1080_v3
33
+
34
  **Dataset Size:** 2-3k
35
+
36
  **OTF (on the fly augmentations):** No
37
+
38
  **Pretrained Model:** 4xESRGAN
39
+
40
  **Iterations:** 140k
41
+
42
  **Batch Size:** 8
43
+
44
  **GT Size:** 64-128
45
 
46
+
47
  **Description:** I gave up on having multiple architectures. This release is ESRGAN only, and provides superior quality compared to AnimeSharpV2 in nearly every scenario. It has most of the advantages of the old V2 Sharp models, while not having issues with depth of field.
48
 
49
+
50
  __Comparisons:__ <https://slow.pics/c/A2BRSa0U>
51
 
52