Conditional-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 Conditional-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 1.25.0	Download
ONNX	w8a16_mixed_fp16	Universal	QAIRT 2.45, ONNX Runtime 1.25.0	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 Conditional-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 Conditional-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: 43.6M
• Model size (float): 166 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® X2 Elite	9.271 ms	208 - 208 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® X Elite	20.456 ms	145 - 145 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	15.015 ms	5 - 447 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® 8 Gen 1 Mobile	38.966 ms	5 - 363 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	20.235 ms	0 - 134 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Qualcomm® QCS8450	38.966 ms	5 - 363 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® 8 Elite Mobile	11.159 ms	3 - 317 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	8.222 ms	3 - 317 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Qualcomm® QCS9075	29.627 ms	5 - 50 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Qualcomm® QCS8750	11.159 ms	3 - 317 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Qualcomm® QCS7181	20.456 ms	145 - 145 MB	NPU
Conditional-DETR-ResNet50	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Gen 3 Mobile	61.376 ms	29 - 496 MB	NPU
Conditional-DETR-ResNet50	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Mobile	50.128 ms	23 - 399 MB	NPU
Conditional-DETR-ResNet50	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Gen 5 Mobile	40.243 ms	22 - 430 MB	NPU
Conditional-DETR-ResNet50	ONNX	w8a16_mixed_fp16	Qualcomm® QCS8750	50.128 ms	23 - 399 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	10.061 ms	5 - 5 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® X Elite	23.446 ms	5 - 5 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	16.969 ms	0 - 401 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	44.743 ms	5 - 348 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8275	98.031 ms	2 - 310 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	23.172 ms	5 - 6 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8450	44.743 ms	5 - 348 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Mobile	11.593 ms	5 - 319 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8295P	34.255 ms	0 - 262 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	8.894 ms	5 - 337 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® SA7255P	98.031 ms	2 - 310 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS9075	36.633 ms	7 - 13 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8750	11.593 ms	5 - 319 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS7181	23.446 ms	5 - 5 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	15.703 ms	0 - 436 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	45.955 ms	0 - 371 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS8275	93.325 ms	0 - 337 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	22.077 ms	0 - 3 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® SA8775P	643.457 ms	0 - 12 MB	CPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® SA8650P	643.457 ms	0 - 12 MB	CPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® SA8255P	643.457 ms	0 - 12 MB	CPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS8450	45.955 ms	0 - 371 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite Mobile	11.9 ms	0 - 341 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® SA8295P	34.62 ms	0 - 289 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	9.257 ms	0 - 355 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® SA7255P	93.325 ms	0 - 337 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS9075	33.33 ms	0 - 93 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS8750	11.9 ms	0 - 341 MB	NPU

License

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

References

• Conditional {DETR} for Fast Training Convergence
• 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.