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 2.42, 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 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:
- Model checkpoint: resnet34-ssd1200
- Input resolution: 1x3x1200x1200
- Number of parameters: 20.0M
- Model size (float): 76.2 MB
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 Mobile | 41.026 ms | 17 - 520 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® X Elite | 91.68 ms | 165 - 165 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 62.565 ms | 17 - 531 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® QCS8550 (Proxy) | 90.146 ms | 0 - 43 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 50.11 ms | 1 - 434 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® QCS9075 | 152.817 ms | 16 - 78 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® QCS8750 | 50.11 ms | 1 - 434 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® QCS7181 | 91.68 ms | 165 - 165 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 52.559 ms | 17 - 558 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® X2 Elite | 62.808 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® X Elite | 129.301 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 84.613 ms | 16 - 603 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8275 | 482.475 ms | 16 - 383 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 125.121 ms | 17 - 19 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8775P | 173.655 ms | 17 - 386 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8650P | 173.655 ms | 17 - 386 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8255P | 173.655 ms | 17 - 386 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 263.003 ms | 3 - 509 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® SA7255P | 482.475 ms | 16 - 383 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8295P | 183.092 ms | 0 - 329 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 67.309 ms | 16 - 392 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS9075 | 193.826 ms | 17 - 35 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8750 | 67.309 ms | 16 - 392 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS7181 | 129.301 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 74.289 ms | 0 - 564 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 108.455 ms | 0 - 547 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS8275 | 514.664 ms | 0 - 378 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 144.203 ms | 0 - 3 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA8775P | 183.746 ms | 1 - 427 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA8650P | 183.746 ms | 1 - 427 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA8255P | 183.746 ms | 1 - 427 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 227.465 ms | 1 - 618 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA7255P | 514.664 ms | 0 - 378 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® SA8295P | 201.704 ms | 0 - 354 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 87.419 ms | 0 - 403 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS9075 | 199.919 ms | 0 - 64 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS8750 | 87.419 ms | 0 - 403 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.