Mask2Former: Optimized for Qualcomm Devices

Mask2Former is a machine learning model that predicts masks and classes of objects in an image.

This is based on the implementation of Mask2Former 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 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® X Elite	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Gen 3 Mobile	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Gen 1 Mobile	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS8550 (Proxy)	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Elite Mobile	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	QAIRT 2.45, ONNX Runtime 1.25.0	Download
PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS9075	QAIRT 2.45, ONNX Runtime 1.25.0	Download
QNN_CONTEXT_BINARY	float	Snapdragon® X2 Elite	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® X Elite	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® 8 Gen 3 Mobile	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® 8 Gen 1 Mobile	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® QCS8550 (Proxy)	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® SA8775P	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® 8 Elite Mobile	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Snapdragon® 8 Elite Gen 5 Mobile	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® SA7255P	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® SA8295P	QAIRT 2.45	Download
QNN_CONTEXT_BINARY	float	Qualcomm® QCS9075	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit Mask2Former 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 Mask2Former on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

• Model checkpoint: facebook/mask2former-swin-tiny-coco-panoptic
• Input resolution: 384x384
• Number of output classes: 100

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Mask2Former	PRECOMPILED_QNN_ONNX	float	Snapdragon® X2 Elite	67.968 ms	10 - 10 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Snapdragon® X Elite	142.678 ms	109 - 109 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Gen 3 Mobile	104.58 ms	12 - 19 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Gen 1 Mobile	234.545 ms	9 - 23 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS8550 (Proxy)	141.063 ms	0 - 113 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS8450	234.545 ms	9 - 23 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	70.63 ms	9 - 21 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS9075	147.549 ms	7 - 12 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Snapdragon® 8 Elite Mobile	88.646 ms	10 - 18 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS8750	88.646 ms	10 - 18 MB	NPU
Mask2Former	PRECOMPILED_QNN_ONNX	float	Qualcomm® QCS7181	142.678 ms	109 - 109 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Snapdragon® X2 Elite	68.019 ms	2 - 2 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Snapdragon® X Elite	142.573 ms	2 - 2 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Snapdragon® 8 Gen 3 Mobile	104.711 ms	4 - 11 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Snapdragon® 8 Gen 1 Mobile	239.722 ms	2 - 11 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Qualcomm® QCS8275	289.576 ms	2 - 10 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Qualcomm® QCS8550 (Proxy)	141.748 ms	2 - 4 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Qualcomm® QCS8450	239.722 ms	2 - 11 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Snapdragon® 8 Elite Gen 5 Mobile	70.644 ms	2 - 11 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Qualcomm® SA7255P	289.576 ms	2 - 10 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Qualcomm® QCS9075	146.653 ms	2 - 8 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Snapdragon® 8 Elite Mobile	88.654 ms	2 - 11 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Qualcomm® SA8295P	180.889 ms	2 - 8 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Qualcomm® QCS8750	88.654 ms	2 - 11 MB	NPU
Mask2Former	QNN_CONTEXT_BINARY	float	Qualcomm® QCS7181	142.573 ms	2 - 2 MB	NPU

License

• The license for the original implementation of Mask2Former can be found here.

References

• Masked-attention Mask Transformer for Universal Image Segmentation
• 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.