Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -48,7 +48,7 @@ Title: Stand As One (Stadium Chant Edition)
|
|
| 48 |
|
| 49 |
---
|
| 50 |
|
| 51 |
-
Despite not being *exactly* good for image models.
|
| 52 |
|
| 53 |
It can be fairly used for *image recognition* for VL captioner models. To increase their pool of styles and detection of details , and the layout and style makes it easier for their segmentation.
|
| 54 |
|
|
|
|
| 48 |
|
| 49 |
---
|
| 50 |
|
| 51 |
+
Despite not being *exactly* good for image generative models (which generate images), the idea is the other end of such chain.
|
| 52 |
|
| 53 |
It can be fairly used for *image recognition* for VL captioner models. To increase their pool of styles and detection of details , and the layout and style makes it easier for their segmentation.
|
| 54 |
|