Downloads · 30 days
0
basketball-rfdetr-s-560
ortizeg/basketball-rfdetr-s-560
basketball-rfdetr-s-560 is a object detection model from ortizeg. Use it when you need objects located in an image. It is set up for onnx. The card lists the license as apache-2.0.
RF-DETR-Small (v2), a real-time transformer detector fine-tuned on the basketball dataset. Its resolution is LOCKED at 560 (the official-eval ONNX export is 560x560; do not rename to 640). No published basketball test…
Access
Public
Updated Jul 30, 2026
Repo size
121 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.onnx121 MB · 100%
At a glance
- Task
- Object Detection
- Library
- onnx
- License
- apache-2.0
- Access
- Public
- Created
- Jul 30, 2026
- Updated
- Jul 30, 2026
- SHA
- 7a4e4aaf
Model card
From the Hugging Face model README
rfdetr-s-560
RF-DETR-Small (v2), a real-time transformer detector fine-tuned on the basketball dataset. Its resolution is LOCKED at 560 (the official-eval ONNX export is 560x560; do not rename to 640). No published basketball test-set score exists for this checkpoint at authoring time.
Metrics
Measured on basketball-player-detection-3 (test split), via the object-detection-eval harness.
| Metric | 5-class | 10-class |
|---|---|---|
| mAP@50:95 | — | — |
| mAP@50 | — | — |
Preprocessing
- Resize:
square(alignment:none) - Normalize:
mean_std - Channel order:
RGB - Input shape:
(3, 560, 560)(float32)
Provenance
- Source repo: https://github.com/roboflow/rf-detr
- Command:
see docs/provenance/training-runs.md#rfdetr-s-560
Usage
This ONNX file is one of the 7-model roster benchmarked in object-detection-eval. Load it through the registry for verified, hash-checked download and the exact preprocessing recipe above:
from object_detection_eval.registry import ModelRegistry, download_weights
registry = ModelRegistry.from_directory("registry")
card = registry.get("rfdetr-s-560")
weights_path = download_weights(card)
- Task
- Object Detection
- Library
- onnx
- License
- apache-2.0
- Created
- Jul 30, 2026
- Updated
- Jul 30, 2026