Video-Text-to-Text
Transformers
Chinese
sign-language-translation
skeleton-based
hyperbolic-geometry
mT5
Instructions to use fiskenai/Geo-Sign with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use fiskenai/Geo-Sign with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("fiskenai/Geo-Sign", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -26,7 +26,11 @@ Edward Fish, Richard Bowden, CVSSP – University of Surrey (arXiv:2506.00129, M
|
|
| 26 |
**Paper** <https://arxiv.org/pdf/2506.00129v1>
|
| 27 |
|
| 28 |
## Code Use
|
| 29 |
-
Download the weights and data labels from the files section of this repo and add them to the github repository <https://github.com/ed-fish/geo-sign>
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
|
| 31 |
## TL;DR
|
| 32 |
Geo-Sign projects pose-based sign-language features into a learnable **Poincaré ball** and aligns them with text embeddings via a geometric contrastive loss.
|
|
|
|
| 26 |
**Paper** <https://arxiv.org/pdf/2506.00129v1>
|
| 27 |
|
| 28 |
## Code Use
|
| 29 |
+
Download the weights and data labels from the files section of this repo and add them to the github repository <https://github.com/ed-fish/geo-sign>.
|
| 30 |
+
Data -> ./Data
|
| 31 |
+
best.pth -> ./checkpoints/best.pth
|
| 32 |
+
pretraining.pth -> ./checkpoints/pretraining.pth
|
| 33 |
+
pretrained_weight/ -> ./pretrained_weight
|
| 34 |
|
| 35 |
## TL;DR
|
| 36 |
Geo-Sign projects pose-based sign-language features into a learnable **Poincaré ball** and aligns them with text embeddings via a geometric contrastive loss.
|