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 2.50, 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 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® 8 Elite Gen 5 For Galaxy Mobile | 31.98 ms | 2 - 277 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 42.665 ms | 1 - 250 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® X2 Elite | 40.254 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® X Elite | 68.287 ms | 28 - 28 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 50.557 ms | 1 - 364 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 112.708 ms | 16 - 303 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 125.756 ms | 16 - 37 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 68.202 ms | 5 - 336 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® QCS8450 | 112.708 ms | 16 - 303 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 114.807 ms | 16 - 36 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 68.287 ms | 28 - 28 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 42.665 ms | 1 - 250 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 45.702 ms | 0 - 289 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 43.697 ms | 0 - 243 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® X2 Elite | 49.971 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® X Elite | 76.175 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 55.414 ms | 0 - 352 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 118.595 ms | 3 - 314 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 128.392 ms | 17 - 36 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 75.56 ms | 17 - 416 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8450 | 118.595 ms | 3 - 314 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 118.694 ms | 17 - 35 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 76.175 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 43.697 ms | 0 - 243 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8295P | 141.144 ms | 0 - 210 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 33.314 ms | 0 - 303 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 39.825 ms | 1 - 267 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 50.411 ms | 0 - 390 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 122.781 ms | 1 - 330 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 121.722 ms | 0 - 63 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 70.36 ms | 1 - 3 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA8775P | 112.957 ms | 1 - 259 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA8650P | 112.957 ms | 1 - 259 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA8255P | 112.957 ms | 1 - 259 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS8450 | 122.781 ms | 1 - 330 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 113.315 ms | 0 - 63 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 39.825 ms | 1 - 267 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA8295P | 125.091 ms | 1 - 211 MB | NPU |
License
- The license for the original implementation of ResNet34-SSD can be found here.
References
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.