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® 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 | 7.008 ms | 24 - 24 MB | NPU |
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 13.051 ms | 23 - 23 MB | NPU |
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 9.839 ms | 16 - 23 MB | NPU |
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 7.664 ms | 13 - 24 MB | NPU |
| BEVFusionDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.736 ms | 16 - 26 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 7.471 ms | 5 - 5 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 13.17 ms | 5 - 5 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 9.88 ms | 5 - 12 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 92.843 ms | 0 - 9 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 7.798 ms | 0 - 9 MB | NPU |
| BEVFusionDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.704 ms | 5 - 14 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 381.124 ms | 102 - 102 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 679.48 ms | 101 - 101 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 537.103 ms | 31 - 43 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 422.174 ms | 38 - 50 MB | NPU |
| BEVFusionEncoder1 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 346.832 ms | 51 - 60 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 385.223 ms | 12 - 12 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 683.999 ms | 12 - 12 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 534.845 ms | 13 - 19 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 1093.502 ms | 1 - 10 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 418.138 ms | 12 - 21 MB | NPU |
| BEVFusionEncoder1 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 348.074 ms | 12 - 22 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 2371.551 ms | 1058 - 1058 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 3435.773 ms | 1058 - 1058 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 2671.903 ms | 589 - 596 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 2438.62 ms | 449 - 456 MB | NPU |
| BEVFusionEncoder2 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2230.688 ms | 382 - 392 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 2409.585 ms | 17 - 17 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 3528.107 ms | 17 - 17 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 2740.678 ms | 17 - 28 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 5600.506 ms | 17 - 26 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 2446.947 ms | 17 - 29 MB | NPU |
| BEVFusionEncoder2 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 2184.698 ms | 17 - 26 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 356.475 ms | 610 - 610 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 700.333 ms | 610 - 610 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 594.808 ms | 576 - 587 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 498.309 ms | 568 - 580 MB | NPU |
| BEVFusionEncoder3 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 370.881 ms | 514 - 523 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 359.01 ms | 610 - 610 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 694.577 ms | 610 - 610 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 588.733 ms | 609 - 620 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 1174.864 ms | 610 - 619 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 498.238 ms | 609 - 622 MB | NPU |
| BEVFusionEncoder3 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 369.237 ms | 609 - 620 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 7.846 ms | 14 - 14 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 12.094 ms | 18 - 18 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 9.303 ms | 33 - 40 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 7.68 ms | 13 - 24 MB | NPU |
| BEVFusionEncoder4 | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 6.78 ms | 31 - 41 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 8.539 ms | 19 - 19 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 12.05 ms | 19 - 19 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 9.156 ms | 19 - 26 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 25.214 ms | 18 - 26 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 7.846 ms | 18 - 27 MB | NPU |
| BEVFusionEncoder4 | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 6.809 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