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 2.42, ONNX Runtime 1.24.1	Download
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a16	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	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® 8 Elite Gen 5 Mobile	8.357 ms	5 - 406 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® X2 Elite	8.912 ms	82 - 82 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® X Elite	19.876 ms	81 - 81 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	14.891 ms	1 - 481 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	19.668 ms	0 - 102 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Qualcomm® QCS9075	30.981 ms	5 - 12 MB	NPU
Conditional-DETR-ResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	10.967 ms	2 - 398 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	8.814 ms	5 - 349 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	10.415 ms	5 - 5 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® X Elite	23.144 ms	5 - 5 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	16.814 ms	4 - 433 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	97.937 ms	0 - 325 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	22.691 ms	5 - 7 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8775P	32.012 ms	1 - 325 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS9075	33.443 ms	5 - 11 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	46.136 ms	4 - 373 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® SA7255P	97.937 ms	0 - 325 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Qualcomm® SA8295P	34.259 ms	0 - 281 MB	NPU
Conditional-DETR-ResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	11.561 ms	5 - 341 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	9.293 ms	0 - 379 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	16.896 ms	0 - 464 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS8275 (Proxy)	92.822 ms	0 - 364 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	22.434 ms	0 - 3 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® SA8775P	30.179 ms	0 - 422 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS9075	33.573 ms	0 - 93 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® QCS8450 (Proxy)	46.043 ms	0 - 404 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® SA7255P	92.822 ms	0 - 364 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Qualcomm® SA8295P	34.37 ms	0 - 311 MB	NPU
Conditional-DETR-ResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	11.885 ms	0 - 375 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.