Update README.md
Browse files
README.md
CHANGED
@@ -11,7 +11,7 @@ widget:
|
|
11 |
example_title: Palace
|
12 |
datasets:
|
13 |
- Colby/autotrain-data-sdxl-detection
|
14 |
-
license: cc-by-
|
15 |
---
|
16 |
|
17 |
# SDXL Detector
|
@@ -22,7 +22,7 @@ This model demonstrates greatly improved performance over the umm-maybe detector
|
|
22 |
|
23 |
However, its performance may be lower for images generated using models other than SDXL. In particular, this model underperforms the original detector for images generated using older models (such as VQGAN+CLIP).
|
24 |
|
25 |
-
|
26 |
|
27 |
# Model Trained Using AutoTrain
|
28 |
|
|
|
11 |
example_title: Palace
|
12 |
datasets:
|
13 |
- Colby/autotrain-data-sdxl-detection
|
14 |
+
license: cc-by-nc-3.0
|
15 |
---
|
16 |
|
17 |
# SDXL Detector
|
|
|
22 |
|
23 |
However, its performance may be lower for images generated using models other than SDXL. In particular, this model underperforms the original detector for images generated using older models (such as VQGAN+CLIP).
|
24 |
|
25 |
+
The data used for this fine-tune is either synthetic (generated by SDXL) and therefore non-copyrightable, or downloaded from Wikimedia and therefore meeting their definition of "free data" (see https://commons.wikimedia.org/wiki/Commons:Licensing for details). However, the original umm-maybe AI art detector was trained on data scraped from image links in Reddit posts, some of which may be copyrighted. Therefore this model as well as its predecessor should be considered appropriate for non-commercial (i.e. personal or educational) fair uses only.
|
26 |
|
27 |
# Model Trained Using AutoTrain
|
28 |
|