Update model card
Browse files
README.md
CHANGED
|
@@ -1,19 +1,6 @@
|
|
| 1 |
-
---
|
| 2 |
-
datasets:
|
| 3 |
-
- cityscapes
|
| 4 |
-
library_name: pytorch
|
| 5 |
-
license: apache-2.0
|
| 6 |
-
pipeline_tag: image-segmentation
|
| 7 |
-
tags:
|
| 8 |
-
- deltatok
|
| 9 |
-
- cvpr2026-highlight
|
| 10 |
-
---
|
| 11 |
-
|
| 12 |
# Segmentation Head — Cityscapes
|
| 13 |
|
| 14 |
-
Segmentation head trained on Cityscapes. Part of [A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens](https://huggingface.co/papers/2604.04913) (CVPR 2026 Highlight).
|
| 15 |
-
|
| 16 |
-
[**Project Page**](https://deltatok.github.io) | [**GitHub**](https://github.com/amazon-far/deltatok) | [**Paper**](https://huggingface.co/papers/2604.04913)
|
| 17 |
|
| 18 |
## Usage
|
| 19 |
|
|
@@ -33,4 +20,4 @@ Requires a frozen [DINOv3](https://github.com/facebookresearch/dinov3) ViT-B bac
|
|
| 33 |
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
|
| 34 |
year = {2026}
|
| 35 |
}
|
| 36 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Segmentation Head — Cityscapes
|
| 2 |
|
| 3 |
+
Segmentation head trained on Cityscapes (mIoU: 70.5). Part of [A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens](https://huggingface.co/papers/2604.04913) (CVPR 2026 Highlight).
|
|
|
|
|
|
|
| 4 |
|
| 5 |
## Usage
|
| 6 |
|
|
|
|
| 20 |
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
|
| 21 |
year = {2026}
|
| 22 |
}
|
| 23 |
+
```
|