DETR-ResNet101-DC5: 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-ResNet101-DC5 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 2.50, ONNX Runtime 1.27.1	Download
QNN_DLC	float	Universal	QAIRT 2.50	Download
TFLITE	float	Universal	QAIRT 2.50	Download

For more device-specific assets and performance metrics, visit DETR-ResNet101-DC5 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-ResNet101-DC5 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Input resolution: 480x480
• Model checkpoint: ResNet101-DC5
• Model size (float): 232 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
DETR-ResNet101-DC5	ONNX	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	19.902 ms	1 - 415 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	25.956 ms	0 - 399 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Snapdragon® X2 Elite	21.536 ms	5 - 5 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Snapdragon® X Elite	48.653 ms	116 - 116 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Snapdragon® 8 Gen 3 Mobile	36.089 ms	1 - 547 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Snapdragon® 8 Gen 1 Mobile	71.23 ms	5 - 457 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	70.712 ms	5 - 13 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	47.794 ms	0 - 126 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Qualcomm® QCS8450	71.23 ms	5 - 457 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	67.145 ms	5 - 13 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	48.653 ms	116 - 116 MB	NPU
DETR-ResNet101-DC5	ONNX	float	Qualcomm® Dragonwing™ Q-8750	25.956 ms	0 - 399 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	24.415 ms	4 - 426 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	28.997 ms	0 - 397 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Snapdragon® X2 Elite	23.996 ms	5 - 5 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Snapdragon® X Elite	53.906 ms	5 - 5 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	36.869 ms	4 - 530 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	69.413 ms	5 - 434 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	73.815 ms	5 - 12 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	53.759 ms	5 - 8 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® SA8775P	70.357 ms	1 - 394 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® SA8650P	70.357 ms	1 - 394 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® SA8255P	70.357 ms	1 - 394 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® QCS8450	69.413 ms	5 - 434 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	70.454 ms	5 - 12 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	53.906 ms	5 - 5 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	28.997 ms	0 - 397 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® SA7255P	213.565 ms	2 - 396 MB	NPU
DETR-ResNet101-DC5	QNN_DLC	float	Qualcomm® SA8295P	75.075 ms	1 - 310 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	24.208 ms	0 - 466 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	28.893 ms	0 - 443 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	36.736 ms	0 - 568 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	69.882 ms	0 - 470 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	73.751 ms	0 - 127 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	50.066 ms	0 - 3 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® SA8775P	70.134 ms	0 - 440 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® SA8650P	70.134 ms	0 - 440 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® SA8255P	70.134 ms	0 - 440 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® QCS8450	69.882 ms	0 - 470 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	70.517 ms	0 - 127 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	28.893 ms	0 - 443 MB	NPU
DETR-ResNet101-DC5	TFLITE	float	Qualcomm® SA8295P	74.542 ms	0 - 349 MB	NPU

License

• The license for the original implementation of DETR-ResNet101-DC5 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.