Instructions to use jozhang97/deta-swin-large-o365 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use jozhang97/deta-swin-large-o365 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("object-detection", model="jozhang97/deta-swin-large-o365")# Load model directly from transformers import AutoModelForObjectDetection model = AutoModelForObjectDetection.from_pretrained("jozhang97/deta-swin-large-o365", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pipeline_tag: object-detection
|
| 3 |
+
tags:
|
| 4 |
+
- vision
|
| 5 |
+
---
|
| 6 |
+
# Detection Transformers with Assignment
|
| 7 |
+
|
| 8 |
+
By [Jeffrey Ouyang-Zhang](https://jozhang97.github.io/), [Jang Hyun Cho](https://sites.google.com/view/janghyuncho/), [Xingyi Zhou](https://www.cs.utexas.edu/~zhouxy/), [Philipp Krähenbühl](http://www.philkr.net/)
|
| 9 |
+
|
| 10 |
+
From the paper [NMS Strikes Back](https://arxiv.org/abs/2212.06137).
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
**TL; DR.** **De**tection **T**ransformers with **A**ssignment (DETA) re-introduce IoU assignment and NMS for transformer-based detectors. DETA trains and tests comparibly as fast as Deformable-DETR and converges much faster (50.2 mAP in 12 epochs on COCO).
|