Objects as Points
Paper • 1904.07850 • Published
CenterNet-3D: Optimized for Qualcomm Devices
CenterNet is a machine learning model for generating a birds eye view represenation from the sensors(cameras) mounted on a vehicle.
This is based on the implementation of CenterNet-3D 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 |
|---|---|---|---|---|
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® X Elite | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® QCS9075 | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® X Elite | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® QCS8550 (Proxy) | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA8775P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA7255P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA8295P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® QCS9075 | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® QCS8450 (Proxy) | QAIRT 2.43 | Download |
For more device-specific assets and performance metrics, visit CenterNet-3D 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 CenterNet-3D on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.driver_assistance
Model Stats:
- Model checkpoint: ddd_3dop.pth
- Input resolution: 1 x 3 x 384 x 1280
- Number of parameters: 20.6M
- Model size: 79 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| CenterNet-3D | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 561.994 ms | 64 - 64 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 944.58 ms | 61 - 61 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 733.967 ms | 8 - 19 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 963.736 ms | 0 - 76 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 962.374 ms | 6 - 14 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 589.066 ms | 3 - 14 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 529.247 ms | 8 - 18 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® X Elite | 2689.409 ms | 55 - 55 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 2313.326 ms | 3 - 10 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 3011.958 ms | 0 - 63 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® QCS9075 | 2872.719 ms | 0 - 6 MB | NPU |
| CenterNet-3D | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2419.424 ms | 1 - 12 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 565.485 ms | 6 - 6 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 947.85 ms | 6 - 6 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 736.644 ms | 8 - 16 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 1462.018 ms | 0 - 9 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 969.606 ms | 6 - 8 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 985.615 ms | 0 - 9 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 961.941 ms | 8 - 17 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | 1007.93 ms | 6 - 16 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 1462.018 ms | 0 - 9 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 1051.635 ms | 0 - 5 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 588.605 ms | 0 - 14 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 532.572 ms | 6 - 16 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® X Elite | 2739.629 ms | 3 - 3 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Gen 3 Mobile | 2438.118 ms | 3 - 11 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® QCS8275 (Proxy) | 3929.358 ms | 1 - 10 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® QCS8550 (Proxy) | 3065.942 ms | 3 - 5 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA8775P | 2970.749 ms | 0 - 9 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® QCS9075 | 2935.006 ms | 5 - 10 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® QCS8450 (Proxy) | 4409.334 ms | 4 - 14 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA7255P | 3929.358 ms | 1 - 10 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA8295P | 3507.321 ms | 0 - 5 MB | NPU |
| CenterNet-3D | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2437.434 ms | 3 - 12 MB | NPU |
License
- The license for the original implementation of CenterNet-3D can be found [here](https://github.com/xingyizhou/CenterNet/blob/master/LICENSE https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf).
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.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support