ResNet34-SSD: Optimized for Qualcomm Devices

ResNet34-SSD is a single-stage object detection model that integrates the ResNet34 backbone with the SSD (Single Shot MultiBox Detector) framework. It is optimized for real-time detection tasks and supports multiple deployment backends including PyTorch, TensorFlow, and ONNX.

This is based on the implementation of ResNet34-SSD 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
QNN_DLC	float	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit ResNet34-SSD 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 ResNet34-SSD on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

Input resolution: 1x3x1200x1200
Model checkpoint: resnet34-ssd1200
Model size (float): 76.2 MB
Number of parameters: 20.0M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ResNet34-SSD	ONNX	float	Snapdragon® X2 Elite	58.12 ms	17 - 17 MB	NPU
ResNet34-SSD	ONNX	float	Snapdragon® X Elite	88.322 ms	30 - 30 MB	NPU
ResNet34-SSD	ONNX	float	Snapdragon® 8 Gen 3 Mobile	63.248 ms	0 - 498 MB	NPU
ResNet34-SSD	ONNX	float	Snapdragon® 8 Gen 1 Mobile	176.14 ms	1 - 431 MB	NPU
ResNet34-SSD	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	142.177 ms	16 - 36 MB	NPU
ResNet34-SSD	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	88.219 ms	0 - 31 MB	NPU
ResNet34-SSD	ONNX	float	Qualcomm® QCS8450	176.14 ms	1 - 431 MB	NPU
ResNet34-SSD	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	134.555 ms	16 - 36 MB	NPU
ResNet34-SSD	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	88.322 ms	30 - 30 MB	NPU
ResNet34-SSD	ONNX	float	Qualcomm® Dragonwing™ Q-8750	51.962 ms	1 - 423 MB	NPU
ResNet34-SSD	ONNX	float	Snapdragon® 8 Elite Mobile	51.962 ms	1 - 423 MB	NPU
ResNet34-SSD	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	39.039 ms	0 - 502 MB	NPU
ResNet34-SSD	QNN_DLC	float	Snapdragon® X2 Elite	62.543 ms	17 - 17 MB	NPU
ResNet34-SSD	QNN_DLC	float	Snapdragon® X Elite	129.06 ms	17 - 17 MB	NPU
ResNet34-SSD	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	85.06 ms	17 - 606 MB	NPU
ResNet34-SSD	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	260.289 ms	7 - 513 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	168.523 ms	17 - 35 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	481.811 ms	16 - 384 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	128.405 ms	17 - 19 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® SA8775P	173.792 ms	16 - 385 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® SA8650P	173.792 ms	16 - 385 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® SA8255P	173.792 ms	16 - 385 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® QCS8450	260.289 ms	7 - 513 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	172.747 ms	18 - 37 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	129.06 ms	17 - 17 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	67.105 ms	0 - 375 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® SA7255P	481.811 ms	16 - 384 MB	NPU
ResNet34-SSD	QNN_DLC	float	Qualcomm® SA8295P	184.165 ms	0 - 328 MB	NPU
ResNet34-SSD	QNN_DLC	float	Snapdragon® 8 Elite Mobile	67.105 ms	0 - 375 MB	NPU
ResNet34-SSD	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	52.526 ms	0 - 546 MB	NPU
ResNet34-SSD	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	107.13 ms	0 - 545 MB	NPU
ResNet34-SSD	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	230.418 ms	1 - 621 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	187.347 ms	0 - 66 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	513.379 ms	0 - 378 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	145.071 ms	0 - 3 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® SA8775P	183.745 ms	1 - 426 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® SA8650P	183.745 ms	1 - 426 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® SA8255P	183.745 ms	1 - 426 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® QCS8450	230.418 ms	1 - 621 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	209.278 ms	0 - 65 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	87.974 ms	0 - 402 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® SA7255P	513.379 ms	0 - 378 MB	NPU
ResNet34-SSD	TFLITE	float	Qualcomm® SA8295P	201.908 ms	0 - 354 MB	NPU
ResNet34-SSD	TFLITE	float	Snapdragon® 8 Elite Mobile	87.974 ms	0 - 402 MB	NPU
ResNet34-SSD	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	72.044 ms	0 - 570 MB	NPU

License

The license for the original implementation of ResNet34-SSD can be found here.

References

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.

Downloads last month: -; Downloads are not tracked for this model. How to track