Update README.md
Browse files
README.md
CHANGED
|
@@ -17,12 +17,13 @@ The collected clean HSIs are averaged, and the paired data are manually aligned
|
|
| 17 |
We randomly select 273 pairs (91 scenes) for training and the remaining 30 (10 scenes) for testing.
|
| 18 |
|
| 19 |
## Train
|
| 20 |
-
We crop overlapped \\(128\times128\\) spatial regions from the paired
|
| 21 |
-
data for training.
|
| 22 |
|
| 23 |
## Inference
|
| 24 |
-
We crop center \\(512\times512\\) spatial region from the paired
|
| 25 |
-
|
|
|
|
|
|
|
| 26 |
|
| 27 |
---
|
| 28 |
license: cc-by-4.0
|
|
|
|
| 17 |
We randomly select 273 pairs (91 scenes) for training and the remaining 30 (10 scenes) for testing.
|
| 18 |
|
| 19 |
## Train
|
| 20 |
+
We crop overlapped \\(128\times128\\) spatial regions from the paired data for training.
|
|
|
|
| 21 |
|
| 22 |
## Inference
|
| 23 |
+
We crop center \\(512\times512\\) spatial region from the paired data for test.
|
| 24 |
+
|
| 25 |
+
## Process
|
| 26 |
+
You can follow [SERT](https://github.com/MyuLi/SERT) to load HSIs.
|
| 27 |
|
| 28 |
---
|
| 29 |
license: cc-by-4.0
|