Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,35 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: gpl-3.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gpl-3.0
|
| 3 |
+
library_name: onnxruntime
|
| 4 |
+
tags:
|
| 5 |
+
- onnx
|
| 6 |
+
- text-detection
|
| 7 |
+
- computer-vision
|
| 8 |
+
- japanese
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# meiki.text.detect.v0
|
| 12 |
+
|
| 13 |
+
experimental text detection models with focus on low latency. trained on japanese video games and manga.
|
| 14 |
+
|
| 15 |
+
model versions:
|
| 16 |
+
- tiny: good for images with only few textlines (e.g. visual novels). ~30ms latency on CPU. ~3ms on GPU.
|
| 17 |
+
- small: better for cases with many textlines (e.g. manga). ~70ms latency on CPU. ~7ms on GPU.
|
| 18 |
+
|
| 19 |
+
fine-tune of https://github.com/Peterande/D-FINE
|
| 20 |
+
|
| 21 |
+
## examples
|
| 22 |
+
|
| 23 |
+
### visual novel
|
| 24 |
+
|
| 25 |
+
| small | tiny |
|
| 26 |
+
|:---:|:---:|
|
| 27 |
+
|  |  |
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
### manga
|
| 31 |
+
|
| 32 |
+
| small | tiny |
|
| 33 |
+
|:---:|:---:|
|
| 34 |
+
|  |  |
|
| 35 |
+
|