Instructions to use HanClinto/cornelius with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- CollectorVision
How to use HanClinto/cornelius with CollectorVision:
pip install git+https://github.com/HanClinto/CollectorVision huggingface_hub
from huggingface_hub import hf_hub_download import collector_vision as cvg checkpoint = hf_hub_download(repo_id="HanClinto/cornelius", filename="model.onnx") # Detector models, such as Cornelius: detector = cvg.NeuralCornerDetector(checkpoint) # Embedder models, such as Milo: embedder = cvg.NeuralEmbedder(checkpoint)
- Notebooks
- Google Colab
- Kaggle
Document fastweb-single stable availability
Browse files
README.md
CHANGED
|
@@ -51,8 +51,9 @@ degenerate predictions where two predicted corners nearly coincide.
|
|
| 51 |
|
| 52 |
### Cornelius vs fastweb-single
|
| 53 |
|
| 54 |
-
Cornelius remains the stable, conservative default. The `fastweb-single`
|
| 55 |
-
|
|
|
|
| 56 |
evaluated as a faster replacement candidate. Both models use the same 384×384
|
| 57 |
input contract and produce `corners`, compatibility `presence`, and `sharpness`
|
| 58 |
outputs.
|
|
@@ -60,7 +61,7 @@ outputs.
|
|
| 60 |
| Model | Channel | Normal test IoU | Normal collapse | Full ±50° mean / p10 IoU | Full-rotation bad / collapse | ONNX size | Params | Local CPU speed |
|
| 61 |
|---|---|---:|---:|---:|---:|---:|---:|---:|
|
| 62 |
| Cornelius 2.12 | stable | **0.968911** | 18 | 0.970411 / 0.954621 | **0 / 0** | 4.41 MB | 1.05M | 23.2 ms / 43.0 FPS |
|
| 63 |
-
| fastweb-single 1.39 | testing | 0.968839 | **13** | **0.974255 / 0.960757** | 1 / 0 | **3.19 MB** | **0.77M** | **8.4 ms / 119.4 FPS** |
|
| 64 |
|
| 65 |
Takeaways:
|
| 66 |
|
|
|
|
| 51 |
|
| 52 |
### Cornelius vs fastweb-single
|
| 53 |
|
| 54 |
+
Cornelius remains the stable, conservative default. The `fastweb-single`
|
| 55 |
+
family is also available through the stable channel for consumers that explicitly
|
| 56 |
+
select it; it is a smaller EfficientViT-B0 global-token SimCC detector being
|
| 57 |
evaluated as a faster replacement candidate. Both models use the same 384×384
|
| 58 |
input contract and produce `corners`, compatibility `presence`, and `sharpness`
|
| 59 |
outputs.
|
|
|
|
| 61 |
| Model | Channel | Normal test IoU | Normal collapse | Full ±50° mean / p10 IoU | Full-rotation bad / collapse | ONNX size | Params | Local CPU speed |
|
| 62 |
|---|---|---:|---:|---:|---:|---:|---:|---:|
|
| 63 |
| Cornelius 2.12 | stable | **0.968911** | 18 | 0.970411 / 0.954621 | **0 / 0** | 4.41 MB | 1.05M | 23.2 ms / 43.0 FPS |
|
| 64 |
+
| fastweb-single 1.39 | stable + testing | 0.968839 | **13** | **0.974255 / 0.960757** | 1 / 0 | **3.19 MB** | **0.77M** | **8.4 ms / 119.4 FPS** |
|
| 65 |
|
| 66 |
Takeaways:
|
| 67 |
|