DETR-ResNet50: Optimized for Qualcomm Devices

DETR is a machine learning model that can detect objects (trained on COCO dataset).

This is based on the implementation of DETR-ResNet50 found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime	Precision	Chipset	SDK Versions	Download
ONNX	float	Universal	QAIRT 2.45, ONNX Runtime 1.27.1	Download
ONNX	w8a16_mixed_fp16	Universal	QAIRT 2.45, ONNX Runtime 1.27.1	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit DETR-ResNet50 on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

• Custom weights (e.g., fine-tuned checkpoints)
• Custom input shapes
• Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for DETR-ResNet50 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Input resolution: 480x480
• Model checkpoint: ResNet50
• Model size (float): 158 MB
• Number of parameters: 41.4M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
DETR-ResNet50	ONNX	float	Snapdragon® X2 Elite	11.526 ms	5 - 5 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® X Elite	17.815 ms	78 - 78 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	13.383 ms	0 - 369 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® 8 Gen 1 Mobile	35.706 ms	1 - 318 MB	NPU
DETR-ResNet50	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	28.837 ms	5 - 13 MB	NPU
DETR-ResNet50	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	18.011 ms	0 - 455 MB	NPU
DETR-ResNet50	ONNX	float	Qualcomm® QCS8450	35.706 ms	1 - 318 MB	NPU
DETR-ResNet50	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	27.268 ms	5 - 12 MB	NPU
DETR-ResNet50	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	17.815 ms	78 - 78 MB	NPU
DETR-ResNet50	ONNX	float	Qualcomm® Dragonwing™ Q-8750	9.98 ms	2 - 273 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® 8 Elite Mobile	9.98 ms	2 - 273 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	7.448 ms	3 - 317 MB	NPU
DETR-ResNet50	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Gen 3 Mobile	39.8 ms	13 - 355 MB	NPU
DETR-ResNet50	ONNX	w8a16_mixed_fp16	Qualcomm® Dragonwing™ IQ-8275	68.309 ms	16 - 19 MB	NPU
DETR-ResNet50	ONNX	w8a16_mixed_fp16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	50.827 ms	0 - 50 MB	NPU
DETR-ResNet50	ONNX	w8a16_mixed_fp16	Qualcomm® QCS8450	97.068 ms	7 - 466 MB	NPU
DETR-ResNet50	ONNX	w8a16_mixed_fp16	Qualcomm® Dragonwing™ IQ-9075	51.04 ms	16 - 19 MB	NPU
DETR-ResNet50	ONNX	w8a16_mixed_fp16	Qualcomm® Dragonwing™ Q-8750	33.763 ms	15 - 274 MB	NPU
DETR-ResNet50	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Mobile	33.763 ms	15 - 274 MB	NPU
DETR-ResNet50	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Gen 5 Mobile	28.241 ms	15 - 282 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	8.959 ms	5 - 5 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® X Elite	19.67 ms	5 - 5 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	13.614 ms	3 - 357 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	41.581 ms	5 - 304 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	27.761 ms	5 - 12 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	86.027 ms	1 - 268 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	19.229 ms	5 - 314 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8775P	26.888 ms	0 - 300 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8650P	26.888 ms	0 - 300 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8255P	26.888 ms	0 - 300 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8450	41.581 ms	5 - 304 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	27.282 ms	5 - 11 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	19.67 ms	5 - 5 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	10.472 ms	0 - 275 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA7255P	86.027 ms	1 - 268 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8295P	31.197 ms	0 - 224 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Mobile	10.472 ms	0 - 275 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	7.895 ms	5 - 288 MB	NPU
DETR-ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	13.686 ms	0 - 391 MB	NPU
DETR-ResNet50	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	41.613 ms	0 - 328 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	27.586 ms	0 - 89 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	85.963 ms	0 - 294 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	19.037 ms	0 - 2 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA8775P	26.901 ms	0 - 323 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA8650P	26.901 ms	0 - 323 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA8255P	26.901 ms	0 - 323 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® QCS8450	41.613 ms	0 - 328 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	27.49 ms	0 - 88 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	10.509 ms	0 - 306 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA7255P	85.963 ms	0 - 294 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA8295P	31.095 ms	0 - 244 MB	NPU
DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite Mobile	10.509 ms	0 - 306 MB	NPU
DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	7.99 ms	0 - 312 MB	NPU

License

• The license for the original implementation of DETR-ResNet50 can be found here.

References

• End-to-End Object Detection with Transformers
• Source Model Implementation

Community

• Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
• For questions or feedback please reach out to us.