Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,17 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
+
## Third-party Baselines Used in the Demo
|
| 5 |
+
|
| 6 |
+
The accompanying CapCLIP Hugging Face Space includes the following third-party baseline models for comparison:
|
| 7 |
+
|
| 8 |
+
| Model | Source / Authors | Use in Demo |
|
| 9 |
+
|---|---|---|
|
| 10 |
+
| MedCLIP | Wang et al., EMNLP 2022 | Medical vision-language baseline |
|
| 11 |
+
| BioMedCLIP | Microsoft Research | Biomedical vision-language baseline |
|
| 12 |
+
| MedImageInsight | Codella et al. / public Hugging Face implementation | General medical imaging embedding baseline |
|
| 13 |
+
| OpenAI CLIP | OpenAI | General-domain CLIP baseline |
|
| 14 |
+
|
| 15 |
+
These models are not proposed as part of CapCLIP. They are included to allow transparent comparison with general-domain, biomedical, and medical imaging foundation models. We acknowledge and thank the original authors and organisations for releasing their code and/or model weights.
|
| 16 |
+
|
| 17 |
+
Users should consult the original model cards, repositories, licences, and intended-use statements before reusing any third-party baseline model outside this research demo.
|