Hugging Face's logo

qualcomm
/
Segformer-Base

Image Segmentation
PyTorch
LiteRT
real_time
android
Model card Files
xet
Community

Segformer-Base: Optimized for Mobile Deployment

Real-time object segmentation

Segformer Base is a machine learning model that predicts masks and classes of objects in an image.

This model is an implementation of Segformer-Base found here.

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

Model Details

  • Model Type: Model_use_case.semantic_segmentation
  • Model Stats:
    • Model checkpoint: nvidia/segformer-b0-finetuned-ade-512-512
    • Input resolution: 512x512
    • Number of output classes: 150
    • Number of parameters: 3.75M
    • Model size (float): 14.4 MB
    • Model size (w8a16): 4.57 MB
    • Model size (w8a8): 3.90 MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
Segformer-Base float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 216.605 ms 8 - 57 MB NPU Segformer-Base.tflite
Segformer-Base float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 210.564 ms 2 - 43 MB NPU Segformer-Base.dlc
Segformer-Base float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 123.972 ms 9 - 82 MB NPU Segformer-Base.tflite
Segformer-Base float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 116.901 ms 3 - 56 MB NPU Segformer-Base.dlc
Segformer-Base float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 110.63 ms 10 - 28 MB NPU Segformer-Base.tflite
Segformer-Base float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 108.135 ms 3 - 19 MB NPU Segformer-Base.dlc
Segformer-Base float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 109.716 ms 19 - 50 MB NPU Segformer-Base.onnx.zip
Segformer-Base float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 113.257 ms 9 - 58 MB NPU Segformer-Base.tflite
Segformer-Base float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 110.23 ms 2 - 43 MB NPU Segformer-Base.dlc
Segformer-Base float SA7255P ADP Qualcomm® SA7255P TFLITE 216.605 ms 8 - 57 MB NPU Segformer-Base.tflite
Segformer-Base float SA7255P ADP Qualcomm® SA7255P QNN_DLC 210.564 ms 2 - 43 MB NPU Segformer-Base.dlc
Segformer-Base float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 111.018 ms 10 - 30 MB NPU Segformer-Base.tflite
Segformer-Base float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 108.158 ms 3 - 19 MB NPU Segformer-Base.dlc
Segformer-Base float SA8295P ADP Qualcomm® SA8295P TFLITE 136.293 ms 9 - 80 MB NPU Segformer-Base.tflite
Segformer-Base float SA8295P ADP Qualcomm® SA8295P QNN_DLC 119.953 ms 2 - 51 MB NPU Segformer-Base.dlc
Segformer-Base float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 110.688 ms 9 - 26 MB NPU Segformer-Base.tflite
Segformer-Base float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 108.405 ms 3 - 18 MB NPU Segformer-Base.dlc
Segformer-Base float SA8775P ADP Qualcomm® SA8775P TFLITE 113.257 ms 9 - 58 MB NPU Segformer-Base.tflite
Segformer-Base float SA8775P ADP Qualcomm® SA8775P QNN_DLC 110.23 ms 2 - 43 MB NPU Segformer-Base.dlc
Segformer-Base float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 82.893 ms 8 - 67 MB NPU Segformer-Base.tflite
Segformer-Base float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 82.137 ms 3 - 52 MB NPU Segformer-Base.dlc
Segformer-Base float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 82.757 ms 22 - 77 MB NPU Segformer-Base.onnx.zip
Segformer-Base float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 74.565 ms 8 - 63 MB NPU Segformer-Base.tflite
Segformer-Base float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 73.303 ms 3 - 50 MB NPU Segformer-Base.dlc
Segformer-Base float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 76.69 ms 21 - 66 MB NPU Segformer-Base.onnx.zip
Segformer-Base float Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 113.206 ms 3 - 3 MB NPU Segformer-Base.dlc
Segformer-Base float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 115.388 ms 33 - 33 MB NPU Segformer-Base.onnx.zip
Segformer-Base w8a16 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 27.083 ms 1 - 44 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 20.507 ms 2 - 50 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 15.768 ms 1 - 16 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 16.142 ms 2 - 40 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) QNN_DLC 55.524 ms 2 - 83 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) ONNX 450.669 ms 375 - 390 MB CPU Segformer-Base.onnx.zip
Segformer-Base w8a16 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 382.425 ms 368 - 376 MB CPU Segformer-Base.onnx.zip
Segformer-Base w8a16 SA7255P ADP Qualcomm® SA7255P QNN_DLC 27.083 ms 1 - 44 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 15.738 ms 1 - 15 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 SA8295P ADP Qualcomm® SA8295P QNN_DLC 19.454 ms 2 - 51 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 15.778 ms 1 - 16 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 SA8775P ADP Qualcomm® SA8775P QNN_DLC 16.142 ms 2 - 40 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 10.325 ms 2 - 54 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 8.058 ms 2 - 53 MB NPU Segformer-Base.dlc
Segformer-Base w8a16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 35.7 ms 38 - 521 MB NPU Segformer-Base.onnx.zip
Segformer-Base w8a16 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 16.198 ms 2 - 2 MB NPU Segformer-Base.dlc
Segformer-Base w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 21.325 ms 2 - 39 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 12.235 ms 2 - 49 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 11.759 ms 2 - 15 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 64.493 ms 6 - 101 MB NPU Segformer-Base.onnx.zip
Segformer-Base w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 12.369 ms 2 - 40 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) TFLITE 137.92 ms 15 - 52 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) ONNX 273.902 ms 227 - 243 MB CPU Segformer-Base.onnx.zip
Segformer-Base w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) TFLITE 414.071 ms 4 - 42 MB CPU Segformer-Base.tflite
Segformer-Base w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 237.072 ms 226 - 238 MB CPU Segformer-Base.onnx.zip
Segformer-Base w8a8 SA7255P ADP Qualcomm® SA7255P TFLITE 21.325 ms 2 - 39 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 11.777 ms 2 - 14 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 SA8295P ADP Qualcomm® SA8295P TFLITE 14.376 ms 2 - 47 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 11.773 ms 2 - 15 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 SA8775P ADP Qualcomm® SA8775P TFLITE 12.369 ms 2 - 40 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 8.159 ms 1 - 48 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 49.693 ms 13 - 241 MB NPU Segformer-Base.onnx.zip
Segformer-Base w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 6.765 ms 2 - 45 MB NPU Segformer-Base.tflite
Segformer-Base w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 33.557 ms 24 - 242 MB NPU Segformer-Base.onnx.zip
Segformer-Base w8a8 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 65.176 ms 29 - 29 MB NPU Segformer-Base.onnx.zip

Installation

Install the package via pip:

pip install qai-hub-models

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

Sign-in to Qualcomm® AI Hub 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.segformer_base.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.segformer_base.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.segformer_base.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.segformer_base 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. Sign up for access.

Run demo on a cloud-hosted device

You can also run the demo on-device.

python -m qai_hub_models.models.segformer_base.demo --eval-mode on-device

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.segformer_base.demo -- --eval-mode on-device

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 Segformer-Base's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

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

References

Community

Downloads last month
202
Inference Providers NEW
Image Segmentation
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support