BEVFusion: Optimized for Mobile Deployment

Construct a bird’s eye view from sensors mounted on a vehicle

BeVFusion is a machine learning model for generating a birds eye view represenation from the sensors(cameras) mounted on a vehicle.

This model is an implementation of BEVFusion found here.

This repository provides scripts to run BEVFusion on Qualcomm® devices. More details on model performance across various devices, can be found here.

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

Model	Precision	Device	Chipset	Target Runtime	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit	Target Model
BEVFusionEncoder1	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	PRECOMPILED_QNN_ONNX	851.341 ms	0 - 100 MB	NPU	Use Export Script
BEVFusionEncoder1	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_CONTEXT_BINARY	708.316 ms	1 - 11 MB	NPU	Use Export Script
BEVFusionEncoder1	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_CONTEXT_BINARY	535.84 ms	13 - 31 MB	NPU	Use Export Script
BEVFusionEncoder1	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	PRECOMPILED_QNN_ONNX	618.606 ms	31 - 46 MB	NPU	Use Export Script
BEVFusionEncoder1	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_CONTEXT_BINARY	427.006 ms	12 - 26 MB	NPU	Use Export Script
BEVFusionEncoder1	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	PRECOMPILED_QNN_ONNX	506.321 ms	19 - 34 MB	NPU	Use Export Script
BEVFusionEncoder1	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_CONTEXT_BINARY	361.104 ms	12 - 23 MB	NPU	Use Export Script
BEVFusionEncoder1	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	PRECOMPILED_QNN_ONNX	437.45 ms	46 - 57 MB	NPU	Use Export Script
BEVFusionEncoder1	float	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_CONTEXT_BINARY	689.862 ms	12 - 12 MB	NPU	Use Export Script
BEVFusionEncoder1	float	Snapdragon X Elite CRD	Snapdragon® X Elite	PRECOMPILED_QNN_ONNX	813.606 ms	94 - 94 MB	NPU	Use Export Script
BEVFusionEncoder2	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	PRECOMPILED_QNN_ONNX	3407.415 ms	587 - 596 MB	NPU	Use Export Script
BEVFusionEncoder2	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_CONTEXT_BINARY	3249.28 ms	16 - 27 MB	NPU	Use Export Script
BEVFusionEncoder2	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_CONTEXT_BINARY	2592.805 ms	17 - 37 MB	NPU	Use Export Script
BEVFusionEncoder2	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	PRECOMPILED_QNN_ONNX	2591.196 ms	596 - 615 MB	NPU	Use Export Script
BEVFusionEncoder2	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_CONTEXT_BINARY	2379.048 ms	17 - 30 MB	NPU	Use Export Script
BEVFusionEncoder2	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	PRECOMPILED_QNN_ONNX	2368.267 ms	474 - 485 MB	NPU	Use Export Script
BEVFusionEncoder2	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_CONTEXT_BINARY	2177.685 ms	17 - 28 MB	NPU	Use Export Script
BEVFusionEncoder2	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	PRECOMPILED_QNN_ONNX	2109.286 ms	49 - 59 MB	NPU	Use Export Script
BEVFusionEncoder2	float	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_CONTEXT_BINARY	3348.536 ms	17 - 17 MB	NPU	Use Export Script
BEVFusionEncoder2	float	Snapdragon X Elite CRD	Snapdragon® X Elite	PRECOMPILED_QNN_ONNX	3252.187 ms	1058 - 1058 MB	NPU	Use Export Script
BEVFusionEncoder3	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	PRECOMPILED_QNN_ONNX	548.211 ms	609 - 612 MB	NPU	Use Export Script
BEVFusionEncoder3	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_CONTEXT_BINARY	719.135 ms	609 - 619 MB	NPU	Use Export Script
BEVFusionEncoder3	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_CONTEXT_BINARY	582.637 ms	609 - 628 MB	NPU	Use Export Script
BEVFusionEncoder3	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	PRECOMPILED_QNN_ONNX	456.969 ms	597 - 617 MB	NPU	Use Export Script
BEVFusionEncoder3	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_CONTEXT_BINARY	503.285 ms	609 - 626 MB	NPU	Use Export Script
BEVFusionEncoder3	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	PRECOMPILED_QNN_ONNX	377.651 ms	590 - 605 MB	NPU	Use Export Script
BEVFusionEncoder3	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_CONTEXT_BINARY	370.888 ms	609 - 624 MB	NPU	Use Export Script
BEVFusionEncoder3	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	PRECOMPILED_QNN_ONNX	331.756 ms	602 - 616 MB	NPU	Use Export Script
BEVFusionEncoder3	float	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_CONTEXT_BINARY	698.865 ms	610 - 610 MB	NPU	Use Export Script
BEVFusionEncoder3	float	Snapdragon X Elite CRD	Snapdragon® X Elite	PRECOMPILED_QNN_ONNX	512.279 ms	610 - 610 MB	NPU	Use Export Script
BEVFusionEncoder4	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	PRECOMPILED_QNN_ONNX	12.891 ms	23 - 28 MB	NPU	Use Export Script
BEVFusionEncoder4	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_CONTEXT_BINARY	40.912 ms	18 - 28 MB	NPU	Use Export Script
BEVFusionEncoder4	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_CONTEXT_BINARY	8.926 ms	18 - 36 MB	NPU	Use Export Script
BEVFusionEncoder4	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	PRECOMPILED_QNN_ONNX	9.475 ms	31 - 51 MB	NPU	Use Export Script
BEVFusionEncoder4	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_CONTEXT_BINARY	7.798 ms	13 - 27 MB	NPU	Use Export Script
BEVFusionEncoder4	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	PRECOMPILED_QNN_ONNX	8.005 ms	12 - 23 MB	NPU	Use Export Script
BEVFusionEncoder4	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_CONTEXT_BINARY	6.841 ms	18 - 30 MB	NPU	Use Export Script
BEVFusionEncoder4	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	PRECOMPILED_QNN_ONNX	7.145 ms	31 - 41 MB	NPU	Use Export Script
BEVFusionEncoder4	float	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_CONTEXT_BINARY	11.806 ms	19 - 19 MB	NPU	Use Export Script
BEVFusionEncoder4	float	Snapdragon X Elite CRD	Snapdragon® X Elite	PRECOMPILED_QNN_ONNX	12.077 ms	19 - 19 MB	NPU	Use Export Script
BEVFusionDecoder	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	PRECOMPILED_QNN_ONNX	12.84 ms	14 - 16 MB	NPU	Use Export Script
BEVFusionDecoder	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_CONTEXT_BINARY	23.67 ms	1 - 11 MB	NPU	Use Export Script
BEVFusionDecoder	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_CONTEXT_BINARY	9.688 ms	5 - 23 MB	NPU	Use Export Script
BEVFusionDecoder	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	PRECOMPILED_QNN_ONNX	9.927 ms	16 - 35 MB	NPU	Use Export Script
BEVFusionDecoder	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_CONTEXT_BINARY	7.672 ms	5 - 19 MB	NPU	Use Export Script
BEVFusionDecoder	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	PRECOMPILED_QNN_ONNX	7.906 ms	12 - 26 MB	NPU	Use Export Script
BEVFusionDecoder	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_CONTEXT_BINARY	5.758 ms	5 - 16 MB	NPU	Use Export Script
BEVFusionDecoder	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	PRECOMPILED_QNN_ONNX	6.204 ms	17 - 28 MB	NPU	Use Export Script
BEVFusionDecoder	float	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_CONTEXT_BINARY	13.035 ms	5 - 5 MB	NPU	Use Export Script
BEVFusionDecoder	float	Snapdragon X Elite CRD	Snapdragon® X Elite	PRECOMPILED_QNN_ONNX	13.265 ms	23 - 23 MB	NPU	Use Export Script

Installation

Install the package via pip:

# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
pip install "qai-hub-models[bevfusion-det]"

Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device

Sign-in to Qualcomm® AI Hub Workbench with your Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.

With this API token, you can configure your client to run models on the cloud hosted devices.

qai-hub configure --api_token API_TOKEN

Navigate to docs for more information.

Demo off target

The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.

python -m qai_hub_models.models.bevfusion_det.demo

The above demo runs a reference implementation of pre-processing, model inference, and post processing.

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.bevfusion_det.demo

Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:

Performance check on-device on a cloud-hosted device
Downloads compiled assets that can be deployed on-device for Android.
Accuracy check between PyTorch and on-device outputs.

python -m qai_hub_models.models.bevfusion_det.export

How does this work?

This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:

Step 1: Compile model for on-device deployment

To compile a PyTorch model for on-device deployment, we first trace the model in memory using the jit.trace and then call the submit_compile_job API.

import torch

import qai_hub as hub
from qai_hub_models.models.bevfusion_det import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S25")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

Step 2: Performance profiling on cloud-hosted device

After compiling models from step 1. Models can be profiled model on-device using the target_model. Note that this scripts runs the model on a device automatically provisioned in the cloud. Once the job is submitted, you can navigate to a provided job URL to view a variety of on-device performance metrics.

profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)

Step 3: Verify on-device accuracy

To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.

input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.

Note: This on-device profiling and inference requires access to Qualcomm® AI Hub Workbench. Sign up for access.

Deploying compiled model to Android

The models can be deployed using multiple runtimes:

TensorFlow Lite (.tflite export): This tutorial provides a guide to deploy the .tflite model in an Android application.
QNN (.so export ): This sample app provides instructions on how to use the .so shared library in an Android application.

View on Qualcomm® AI Hub

Get more details on BEVFusion's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

The license for the original implementation of BEVFusion can be found here.
The license for the compiled assets for on-device deployment 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.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

Other

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support