Hugging Face's logo

qualcomm
/
SINet

Image Segmentation
PyTorch
android
Model card Files
xet
 Community

SINet: Optimized for Qualcomm Devices

SINet is a machine learning model that is designed to segment people from close-up portrait images in real time.

This is based on the implementation of SINet 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
ONNX float Universal QAIRT 2.50, ONNX Runtime 1.30.0 Download
ONNX w8a16 Universal QAIRT 2.50, ONNX Runtime 1.30.0 Download
ONNX w8a8 Universal QAIRT 2.50, ONNX Runtime 1.30.0 Download
QNN_DLC float Universal QAIRT 2.50 Download
QNN_DLC w8a16 Universal QAIRT 2.50 Download
QNN_DLC w8a8 Universal QAIRT 2.50 Download
TFLITE float Universal QAIRT 2.50 Download

For more device-specific assets and performance metrics, visit SINet 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 SINet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

  • Input resolution: 224x224
  • Model checkpoint: SINet.pth
  • Model size (float): 415 KB
  • Model size (w8a8): 241 KB
  • Number of output classes: 2 (foreground / background)
  • Number of parameters: 91.9K

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
SINet ONNX float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 1.181 ms 0 - 44 MB NPU
SINet ONNX float Snapdragon® 8 Elite For Galaxy Mobile 1.327 ms 0 - 38 MB NPU
SINet ONNX float Snapdragon® X2 Elite 1.258 ms 2 - 2 MB NPU
SINet ONNX float Snapdragon® X Elite 2.742 ms 2 - 2 MB NPU
SINet ONNX float Snapdragon® 8 Gen 3 Mobile 1.684 ms 0 - 53 MB NPU
SINet ONNX float Snapdragon® 8 Gen 1 Mobile 2.906 ms 1 - 60 MB NPU
SINet ONNX float Qualcomm® Dragonwing™ IQ-8275 2.537 ms 1 - 5 MB NPU
SINet ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 2.6 ms 0 - 3 MB NPU
SINet ONNX float Qualcomm® QCS8450 2.906 ms 1 - 60 MB NPU
SINet ONNX float Qualcomm® Dragonwing™ IQ-9075 2.873 ms 1 - 5 MB NPU
SINet ONNX float Qualcomm® Dragonwing™ IQ-X7181 2.742 ms 2 - 2 MB NPU
SINet ONNX float Qualcomm® Dragonwing™ Q-8750 1.327 ms 0 - 38 MB NPU
SINet ONNX w8a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 1.272 ms 0 - 54 MB NPU
SINet ONNX w8a16 Snapdragon® 8 Elite For Galaxy Mobile 1.481 ms 0 - 56 MB NPU
SINet ONNX w8a16 Snapdragon® X2 Elite 1.318 ms 1 - 1 MB NPU
SINet ONNX w8a16 Snapdragon® X Elite 2.997 ms 1 - 1 MB NPU
SINet ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 1.856 ms 0 - 66 MB NPU
SINet ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 3.504 ms 0 - 69 MB NPU
SINet ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 2.606 ms 0 - 4 MB NPU
SINet ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 2.839 ms 0 - 9 MB NPU
SINet ONNX w8a16 Qualcomm® QCS8450 3.504 ms 0 - 69 MB NPU
SINet ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 3.072 ms 0 - 3 MB NPU
SINet ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 2.997 ms 1 - 1 MB NPU
SINet ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 1.481 ms 0 - 56 MB NPU
SINet ONNX w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.865 ms 0 - 51 MB NPU
SINet ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.98 ms 0 - 52 MB NPU
SINet ONNX w8a8 Snapdragon® X2 Elite 0.887 ms 1 - 1 MB NPU
SINet ONNX w8a8 Snapdragon® X Elite 1.919 ms 1 - 1 MB NPU
SINet ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 1.271 ms 0 - 63 MB NPU
SINet ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 2.077 ms 0 - 64 MB NPU
SINet ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 1.833 ms 0 - 4 MB NPU
SINet ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.836 ms 0 - 3 MB NPU
SINet ONNX w8a8 Qualcomm® QCS8450 2.077 ms 0 - 64 MB NPU
SINet ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 2.001 ms 0 - 3 MB NPU
SINet ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 1.919 ms 1 - 1 MB NPU
SINet ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 0.98 ms 0 - 52 MB NPU
SINet QNN_DLC float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.716 ms 1 - 39 MB NPU
SINet QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 0.859 ms 1 - 40 MB NPU
SINet QNN_DLC float Snapdragon® X2 Elite 0.989 ms 1 - 1 MB NPU
SINet QNN_DLC float Snapdragon® X Elite 1.98 ms 1 - 1 MB NPU
SINet QNN_DLC float Snapdragon® 8 Gen 3 Mobile 1.169 ms 0 - 48 MB NPU
SINet QNN_DLC float Snapdragon® 8 Gen 1 Mobile 2.163 ms 1 - 57 MB NPU
SINet QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 1.807 ms 1 - 5 MB NPU
SINet QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.746 ms 1 - 2 MB NPU
SINet QNN_DLC float Qualcomm® SA8775P 2.161 ms 1 - 40 MB NPU
SINet QNN_DLC float Qualcomm® SA8650P 2.161 ms 1 - 40 MB NPU
SINet QNN_DLC float Qualcomm® SA8255P 2.161 ms 1 - 40 MB NPU
SINet QNN_DLC float Qualcomm® QCS8450 2.163 ms 1 - 57 MB NPU
SINet QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.998 ms 1 - 4 MB NPU
SINet QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 1.98 ms 1 - 1 MB NPU
SINet QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.859 ms 1 - 40 MB NPU
SINet QNN_DLC float Qualcomm® SA7255P 3.904 ms 1 - 37 MB NPU
SINet QNN_DLC float Qualcomm® SA8295P 2.464 ms 0 - 37 MB NPU
SINet QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.799 ms 0 - 47 MB NPU
SINet QNN_DLC w8a16 Snapdragon® 8 Elite For Galaxy Mobile 0.984 ms 0 - 40 MB NPU
SINet QNN_DLC w8a16 Snapdragon® X2 Elite 1.063 ms 0 - 0 MB NPU
SINet QNN_DLC w8a16 Snapdragon® X Elite 2.177 ms 0 - 0 MB NPU
SINet QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 1.312 ms 0 - 52 MB NPU
SINet QNN_DLC w8a16 Snapdragon® 8 Gen 1 Mobile 2.546 ms 0 - 58 MB NPU
SINet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 1.85 ms 0 - 4 MB NPU
SINet QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.932 ms 0 - 24 MB NPU
SINet QNN_DLC w8a16 Qualcomm® SA8775P 2.285 ms 0 - 45 MB NPU
SINet QNN_DLC w8a16 Qualcomm® SA8650P 2.285 ms 0 - 45 MB NPU
SINet QNN_DLC w8a16 Qualcomm® SA8255P 2.285 ms 0 - 45 MB NPU
SINet QNN_DLC w8a16 Qualcomm® QCS8450 2.546 ms 0 - 58 MB NPU
SINet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 2.131 ms 2 - 5 MB NPU
SINet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 2.177 ms 0 - 0 MB NPU
SINet QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 0.984 ms 0 - 40 MB NPU
SINet QNN_DLC w8a16 Qualcomm® SA8295P 2.989 ms 0 - 41 MB NPU
SINet QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.566 ms 0 - 41 MB NPU
SINet QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.718 ms 0 - 38 MB NPU
SINet QNN_DLC w8a8 Snapdragon® X2 Elite 0.778 ms 0 - 0 MB NPU
SINet QNN_DLC w8a8 Snapdragon® X Elite 1.609 ms 0 - 0 MB NPU
SINet QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 0.957 ms 0 - 49 MB NPU
SINet QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 1.619 ms 0 - 55 MB NPU
SINet QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 1.381 ms 0 - 3 MB NPU
SINet QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.395 ms 0 - 2 MB NPU
SINet QNN_DLC w8a8 Qualcomm® QCS8450 1.619 ms 0 - 55 MB NPU
SINet QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 1.535 ms 0 - 3 MB NPU
SINet QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 1.609 ms 0 - 0 MB NPU
SINet QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 0.718 ms 0 - 38 MB NPU
SINet QNN_DLC w8a8 Qualcomm® SA8295P 2.038 ms 0 - 38 MB NPU
SINet TFLITE float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.788 ms 0 - 35 MB NPU
SINet TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 0.929 ms 0 - 37 MB NPU
SINet TFLITE float Snapdragon® 8 Gen 3 Mobile 1.222 ms 0 - 47 MB NPU
SINet TFLITE float Snapdragon® 8 Gen 1 Mobile 2.292 ms 0 - 58 MB NPU
SINet TFLITE float Qualcomm® Dragonwing™ IQ-8275 2.216 ms 0 - 6 MB NPU
SINet TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.815 ms 0 - 2 MB NPU
SINet TFLITE float Qualcomm® QCS8450 2.292 ms 0 - 58 MB NPU
SINet TFLITE float Qualcomm® Dragonwing™ IQ-9075 2.318 ms 2 - 7 MB NPU
SINet TFLITE float Qualcomm® Dragonwing™ Q-8750 0.929 ms 0 - 37 MB NPU
SINet TFLITE float Qualcomm® SA7255P 4.498 ms 2 - 37 MB NPU
SINet TFLITE float Qualcomm® SA8295P 2.814 ms 0 - 36 MB NPU

License

  • The license for the original implementation of SINet can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Image Segmentation
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for qualcomm/SINet