AOT-GAN: Optimized for Qualcomm Devices

AOT-GAN is a machine learning model that allows to erase and in-paint part of given input image.

This is based on the implementation of AOT-GAN 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 2.45, ONNX Runtime 1.27.1	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

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

Model Details

Model Type: Model_use_case.image_editing

Model Stats:

• Input resolution: 512x512
• Model checkpoint: CelebAHQ
• Model size (float): 58.0 MB
• Number of parameters: 15.2M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
AOT-GAN	ONNX	float	Snapdragon® X2 Elite	54.463 ms	8 - 8 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® X Elite	138.79 ms	32 - 32 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® 8 Gen 3 Mobile	96.129 ms	11 - 720 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® 8 Gen 1 Mobile	208.865 ms	1 - 602 MB	NPU
AOT-GAN	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	188.447 ms	4 - 11 MB	NPU
AOT-GAN	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	136.047 ms	0 - 38 MB	NPU
AOT-GAN	ONNX	float	Qualcomm® QCS8450	208.865 ms	1 - 602 MB	NPU
AOT-GAN	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	167.466 ms	4 - 11 MB	NPU
AOT-GAN	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	138.79 ms	32 - 32 MB	NPU
AOT-GAN	ONNX	float	Qualcomm® Dragonwing™ Q-8750	75.445 ms	7 - 616 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® 8 Elite Mobile	75.445 ms	7 - 616 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	50.95 ms	5 - 511 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® X2 Elite	50.875 ms	4 - 4 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® X Elite	122.232 ms	4 - 4 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	88.71 ms	4 - 695 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	198.446 ms	3 - 607 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	541.058 ms	1 - 543 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	121.742 ms	4 - 689 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA8775P	161.21 ms	1 - 543 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA8650P	161.21 ms	1 - 543 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA8255P	161.21 ms	1 - 543 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® QCS8450	198.446 ms	3 - 607 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	161.658 ms	4 - 13 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	122.232 ms	4 - 4 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	69.625 ms	3 - 574 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA7255P	541.058 ms	1 - 543 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA8295P	178.688 ms	1 - 476 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® 8 Elite Mobile	69.625 ms	3 - 574 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	46.964 ms	3 - 491 MB	NPU
AOT-GAN	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	88.177 ms	0 - 718 MB	NPU
AOT-GAN	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	198.387 ms	3 - 640 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	175.176 ms	3 - 48 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	541.312 ms	3 - 557 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	116.889 ms	3 - 5 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA8775P	161.496 ms	3 - 558 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA8650P	161.496 ms	3 - 558 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA8255P	161.496 ms	3 - 558 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® QCS8450	198.387 ms	3 - 640 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	161.646 ms	3 - 48 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	69.22 ms	2 - 590 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA7255P	541.312 ms	3 - 557 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA8295P	178.712 ms	3 - 494 MB	NPU
AOT-GAN	TFLITE	float	Snapdragon® 8 Elite Mobile	69.22 ms	2 - 590 MB	NPU
AOT-GAN	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	47.202 ms	2 - 512 MB	NPU

License

• The license for the original implementation of AOT-GAN can be found here.

References

• Aggregated Contextual Transformations for High-Resolution Image Inpainting
• 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.