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 2.42, ONNX Runtime 1.24.3	Download
ONNX	w8a16_mixed_int16	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	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:

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

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
DETR-ResNet50	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	7.317 ms	5 - 290 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® 8 Elite Mobile	9.765 ms	3 - 278 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® X2 Elite	8.358 ms	77 - 77 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® X Elite	17.859 ms	77 - 77 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® X Elite	17.859 ms	77 - 77 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	13.32 ms	1 - 397 MB	NPU
DETR-ResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	17.616 ms	0 - 95 MB	NPU
DETR-ResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	9.765 ms	3 - 278 MB	NPU
DETR-ResNet50	ONNX	float	Qualcomm® QCS9075	28.248 ms	5 - 12 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	8.105 ms	5 - 303 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Mobile	10.552 ms	0 - 286 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	9.241 ms	5 - 5 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® X Elite	20.034 ms	5 - 5 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® X Elite	20.034 ms	5 - 5 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	14.089 ms	1 - 375 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	19.754 ms	5 - 40 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8775P	27.285 ms	1 - 306 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8775P	27.285 ms	1 - 306 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8775P	27.285 ms	1 - 306 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA7255P	86.788 ms	0 - 280 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8295P	32.161 ms	0 - 238 MB	NPU
DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	10.552 ms	0 - 286 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS9075	30.573 ms	7 - 13 MB	NPU
DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	42.698 ms	3 - 319 MB	NPU
DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	7.966 ms	0 - 309 MB	NPU
DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite Mobile	10.517 ms	0 - 306 MB	NPU
DETR-ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	13.686 ms	0 - 384 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	19.193 ms	0 - 3 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA8775P	26.928 ms	0 - 323 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA8775P	26.928 ms	0 - 323 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA8775P	26.928 ms	0 - 323 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA7255P	86.056 ms	0 - 294 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® SA8295P	31.071 ms	0 - 248 MB	NPU
DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	10.517 ms	0 - 306 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® QCS9075	30.366 ms	0 - 88 MB	NPU
DETR-ResNet50	TFLITE	float	Qualcomm® QCS8450 (Proxy)	42.012 ms	0 - 327 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.