Update README
Browse files
README.md
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- kornia
|
| 5 |
+
- orientation-estimation
|
| 6 |
+
- feature-detection
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# kornia/orinet
|
| 10 |
+
|
| 11 |
+
Pretrained weights for **OriNet**,
|
| 12 |
+
used by [`kornia.feature.LAFOrienter`](https://kornia.readthedocs.io/en/latest/feature.html).
|
| 13 |
+
|
| 14 |
+
OriNet is a CNN that regresses the dominant gradient orientation for a keypoint patch,
|
| 15 |
+
trained with a Siamese setup to produce consistent orientations across viewpoints.
|
| 16 |
+
Companion to AffNet. ECCV 2018.
|
| 17 |
+
|
| 18 |
+
**Original repo:** [ducha-aiki/affnet](https://github.com/ducha-aiki/affnet)
|
| 19 |
+
|
| 20 |
+
## Weights
|
| 21 |
+
|
| 22 |
+
| File | Description |
|
| 23 |
+
|------|-------------|
|
| 24 |
+
| `OriNet.pth` | Default checkpoint |
|
| 25 |
+
|
| 26 |
+
## Citation
|
| 27 |
+
|
| 28 |
+
```bibtex
|
| 29 |
+
@inproceedings{AffNet2018,
|
| 30 |
+
author = {D. Mishkin and F. Radenovic and J. Matas},
|
| 31 |
+
title = {{Repeatability is Not Enough}: Learning Affine Regions
|
| 32 |
+
via Discriminability},
|
| 33 |
+
booktitle = {ECCV},
|
| 34 |
+
year = {2018}
|
| 35 |
+
}
|
| 36 |
+
```
|