Remove links to private GitHub repo
Browse files
README.md
CHANGED
|
@@ -53,19 +53,15 @@ For context: the original course leaderboard had 28 teams. This release would pl
|
|
| 53 |
|
| 54 |
## Quick start
|
| 55 |
|
| 56 |
-
|
| 57 |
-
git clone https://github.com/enorenio/Challenge
|
| 58 |
-
cd Challenge
|
| 59 |
-
pip install torch numpy pillow opencv-python scipy
|
| 60 |
|
| 61 |
-
|
| 62 |
|
| 63 |
-
|
| 64 |
-
--ckpt-dirs runs/runs_v4:64:44 runs/runs_v7_pseudo:64:47 \
|
| 65 |
-
--gpu 0
|
| 66 |
-
```
|
| 67 |
|
| 68 |
-
|
|
|
|
|
|
|
| 69 |
|
| 70 |
## What is in this repo
|
| 71 |
|
|
@@ -109,8 +105,7 @@ PASCAL VOC domain. Training images are natural indoor and outdoor scenes from PA
|
|
| 109 |
@misc{morshnev2025scribbleseg,
|
| 110 |
author = {Aleksey Morshnev},
|
| 111 |
title = {Scribble Segmentation Ensemble},
|
| 112 |
-
year = {2025}
|
| 113 |
-
url = {https://github.com/enorenio/Challenge}
|
| 114 |
}
|
| 115 |
```
|
| 116 |
|
|
|
|
| 53 |
|
| 54 |
## Quick start
|
| 55 |
|
| 56 |
+
The inference code is not yet public. The weights here are usable once paired with the matching loader, which will be released alongside a writeup.
|
|
|
|
|
|
|
|
|
|
| 57 |
|
| 58 |
+
In the meantime, the interactive demo at https://enorenio.github.io/scribble-seg-demo/ shows side by side predictions for every method on all 682 train and test images, plus an analysis of the five universally hardest cases.
|
| 59 |
|
| 60 |
+
Downloading the weights from this repo:
|
|
|
|
|
|
|
|
|
|
| 61 |
|
| 62 |
+
```bash
|
| 63 |
+
hf download Enorenio/scribble-segmentation --local-dir runs/
|
| 64 |
+
```
|
| 65 |
|
| 66 |
## What is in this repo
|
| 67 |
|
|
|
|
| 105 |
@misc{morshnev2025scribbleseg,
|
| 106 |
author = {Aleksey Morshnev},
|
| 107 |
title = {Scribble Segmentation Ensemble},
|
| 108 |
+
year = {2025}
|
|
|
|
| 109 |
}
|
| 110 |
```
|
| 111 |
|