FCN-ResNet50-Quantized: Optimized for Mobile Deployment

Quantized fully-convolutional network model for image segmentation

FCN_ResNet50 is a quantized machine learning model that can segment images from the COCO dataset. It uses ResNet50 as a backbone.

This model is an implementation of FCN-ResNet50-Quantized found here. This repository provides scripts to run FCN-ResNet50-Quantized on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

Model Type: Semantic segmentation
Model Stats:
- Model checkpoint: COCO_WITH_VOC_LABELS_V1
- Input resolution: 512x512
- Number of parameters: 33.0M
- Model size: 32.2 MB

Device	Chipset	Target Runtime	Inference Time (ms)	Peak Memory Range (MB)	Precision	Primary Compute Unit	Target Model
Samsung Galaxy S23 Ultra (Android 13)	Snapdragon® 8 Gen 2	TFLite	14.137 ms	7 - 57 MB	INT8	NPU	FCN-ResNet50-Quantized.tflite
Samsung Galaxy S23 Ultra (Android 13)	Snapdragon® 8 Gen 2	QNN Model Library	15.266 ms	1 - 9 MB	INT8	NPU	FCN-ResNet50-Quantized.so

Installation

This model can be installed as a Python package via pip.

pip install "qai-hub-models[fcn_resnet50_quantized]"

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.fcn_resnet50_quantized.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.fcn_resnet50_quantized.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.fcn_resnet50_quantized.export

Profile Job summary of FCN-ResNet50-Quantized
--------------------------------------------------
Device: Snapdragon X Elite CRD (11)
Estimated Inference Time: 16.79 ms
Estimated Peak Memory Range: 0.76-0.76 MB
Compute Units: NPU (79) | Total (79)

Run demo on a cloud-hosted device

You can also run the demo on-device.

python -m qai_hub_models.models.fcn_resnet50_quantized.demo --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.fcn_resnet50_quantized.demo -- --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 FCN-ResNet50-Quantized's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

The license for the original implementation of FCN-ResNet50-Quantized 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.