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 1.24.3 | 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 | 38.733 ms | 17 - 513 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite Mobile | 50.177 ms | 2 - 428 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® X2 Elite | 43.419 ms | 30 - 30 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® X Elite | 91.464 ms | 29 - 29 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® X Elite | 91.464 ms | 29 - 29 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 62.899 ms | 0 - 517 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® QCS8550 (Proxy) | 90.698 ms | 0 - 573 MB | NPU |
| ResNet34-SSD | ONNX | float | Qualcomm® QCS9075 | 152.711 ms | 16 - 36 MB | NPU |
| ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 50.177 ms | 2 - 428 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 52.76 ms | 15 - 553 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 66.869 ms | 16 - 391 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® X2 Elite | 61.836 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® X Elite | 128.96 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® X Elite | 128.96 ms | 17 - 17 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 84.568 ms | 15 - 606 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 481.914 ms | 16 - 385 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 128.396 ms | 17 - 19 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS9075 | 193.951 ms | 17 - 35 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 262.739 ms | 3 - 511 MB | NPU |
| ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 66.869 ms | 16 - 391 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 76.117 ms | 0 - 564 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Mobile | 88.158 ms | 0 - 402 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 107.7 ms | 0 - 543 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 513.324 ms | 0 - 378 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 146.036 ms | 0 - 3 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS9075 | 199.375 ms | 0 - 64 MB | NPU |
| ResNet34-SSD | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 234.949 ms | 1 - 617 MB | NPU |
| ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 88.158 ms | 0 - 402 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.