---
license: apache-2.0
tags:
  - object-detection
  - license-plate-detection
  - rf-detr
  - tensorrt
  - onnx
  - int8
  - edge-deployment
datasets:
  - custom
pipeline_tag: object-detection
---

# RF-DETR License Plate Detector

A fine-tuned RF-DETR Medium model for license plate detection, optimized for edge deployment.

## Model Details

- **Base Model:** RF-DETR Medium
- **Task:** License plate detection (single class)
- **Input Resolution:** 576x576
- **Training Framework:** PyTorch

## Available Formats

| File | Format | Size | Use Case |
|------|--------|------|----------|
| `rfdetr_alpr.onnx` | ONNX | 117MB | Cross-platform inference |
| `rfdetr_alpr_optimized.onnx` | ONNX (optimized) | 115MB | Optimized for TensorRT |
| `rfdetr_alpr_int8.onnx` | ONNX INT8 | 34MB | Edge inference (ARM/NPU) |
| `license_plate_detector_fp16.engine` | TensorRT FP16 | 64MB | GPU inference (balanced) |
| `license_plate_detector_int8.engine` | TensorRT INT8 | 82MB | GPU inference (fastest) |
| `calibration.cache` | Calibration data | 103KB | INT8 calibration cache |

## Deployment Paths

- **NVIDIA GPU**: Use TensorRT engines (`.engine`) for fastest inference
- **Edge/ARM (i.MX8M Plus, i.MX93)**: Use INT8 ONNX with ONNX Runtime or convert to TFLite

## TensorRT Engine Details

- **TensorRT Version:** 10.14.1
- **Target GPU:** NVIDIA GB10 (Compute Capability 12.1)
- **Input Shape:** 1x3x576x576 (fixed batch size)
- **Precision:** FP16 / INT8

## Usage

### ONNX Inference

```python
import onnxruntime as ort
import numpy as np

session = ort.InferenceSession("rfdetr_alpr_optimized.onnx")
# Input: (1, 3, 576, 576) normalized to [0, 1]
outputs = session.run(None, {"images": input_tensor})
boxes, scores = outputs[0], outputs[1]
```

### TensorRT Inference

```python
import tensorrt as trt
import pycuda.driver as cuda

# Load engine
with open("license_plate_detector_int8.engine", "rb") as f:
    engine = trt.Runtime(trt.Logger()).deserialize_cuda_engine(f.read())
```

### Edge Inference (INT8 ONNX)

For ARM/edge devices without NVIDIA GPU:

```python
import onnxruntime as ort

# Use INT8 quantized model for edge deployment
session = ort.InferenceSession(
    "rfdetr_alpr_int8.onnx",
    providers=['CPUExecutionProvider']  # or platform-specific NPU provider
)

# Input: (1, 3, 576, 576) with ImageNet normalization
outputs = session.run(None, {"images": input_tensor})
boxes, scores = outputs[0], outputs[1]
```

## License

Apache 2.0 (same as RF-DETR)