BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation
Paper
• 2205.13542 • Published
BEVFusion: Optimized for Qualcomm Devices
BeVFusion 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 BEVFusion 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 | 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 |
| 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® QCS8275 (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 |
For more device-specific assets and performance metrics, visit BEVFusion 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 BEVFusion on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.driver_assistance
Model Stats:
- Model checkpoint: camera-only-det.pth
- Input resolution: 1 x 6 x 3 x 256 x 704
- Number of parameters: 44M
- Model size: 171 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 7.024 ms | 24 - 24 MB | NPU |
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 13.172 ms | 23 - 23 MB | NPU |
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 9.876 ms | 16 - 23 MB | NPU |
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 7.776 ms | 11 - 22 MB | NPU |
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.742 ms | 18 - 28 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 7.46 ms | 5 - 5 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 13.195 ms | 5 - 5 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 9.87 ms | 5 - 12 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 92.863 ms | 1 - 9 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 7.752 ms | 0 - 13 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.722 ms | 5 - 14 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 379.084 ms | 102 - 102 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 688.815 ms | 101 - 101 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 520.981 ms | 49 - 61 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 417.438 ms | 37 - 50 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 346.26 ms | 50 - 60 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 380.244 ms | 12 - 12 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 686.852 ms | 12 - 12 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 533.864 ms | 13 - 20 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 1096.823 ms | 3 - 10 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 420.597 ms | 12 - 25 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 348.562 ms | 13 - 23 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 2365.465 ms | 1058 - 1058 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 3423.239 ms | 1058 - 1058 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 2723.631 ms | 516 - 522 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 2455.607 ms | 592 - 603 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2198.236 ms | 352 - 362 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 2419.649 ms | 17 - 17 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 3529.497 ms | 17 - 17 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 2702.081 ms | 17 - 27 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 5600.369 ms | 1 - 10 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 2452.24 ms | 17 - 30 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 2198.58 ms | 17 - 26 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 356.884 ms | 610 - 610 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 694.699 ms | 610 - 610 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 593.182 ms | 596 - 608 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 497.271 ms | 585 - 596 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 370.626 ms | 511 - 521 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 358.597 ms | 610 - 610 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 699.571 ms | 610 - 610 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 585.962 ms | 609 - 621 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 1167.477 ms | 611 - 620 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 497.35 ms | 607 - 621 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 368.721 ms | 609 - 620 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 7.808 ms | 13 - 13 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 11.996 ms | 18 - 18 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 9.162 ms | 31 - 37 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 7.748 ms | 11 - 23 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 6.828 ms | 33 - 43 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 8.287 ms | 19 - 19 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 11.954 ms | 19 - 19 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 9.01 ms | 19 - 25 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 25.188 ms | 18 - 27 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 7.741 ms | 17 - 30 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 6.783 ms | 18 - 28 MB | NPU |
License
- The license for the original implementation of BEVFusion can be found here.
References
- BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation
- 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.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support