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 2.42, ONNX Runtime 1.24.3	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:

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

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
AOT-GAN	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	53.714 ms	10 - 508 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® 8 Elite Mobile	74.957 ms	7 - 625 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® X2 Elite	56.727 ms	32 - 32 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® X Elite	138.242 ms	31 - 31 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® X Elite	138.242 ms	31 - 31 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® 8 Gen 3 Mobile	94.378 ms	1 - 746 MB	NPU
AOT-GAN	ONNX	float	Qualcomm® QCS8550 (Proxy)	134.479 ms	0 - 39 MB	NPU
AOT-GAN	ONNX	float	Qualcomm® QCS9075	208.146 ms	4 - 11 MB	NPU
AOT-GAN	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	74.957 ms	7 - 625 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	47.145 ms	6 - 488 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® 8 Elite Mobile	69.595 ms	1 - 573 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® X2 Elite	50.948 ms	4 - 4 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® X Elite	122.164 ms	4 - 4 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® X Elite	122.164 ms	4 - 4 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	88.83 ms	1 - 687 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	541.492 ms	1 - 544 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	120.19 ms	4 - 7 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA8775P	161.325 ms	1 - 543 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA8775P	161.325 ms	1 - 543 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA8775P	161.325 ms	1 - 543 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® QCS9075	211.265 ms	6 - 15 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	198.833 ms	0 - 604 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA7255P	541.492 ms	1 - 544 MB	NPU
AOT-GAN	QNN_DLC	float	Qualcomm® SA8295P	178.727 ms	1 - 474 MB	NPU
AOT-GAN	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	69.595 ms	1 - 573 MB	NPU
AOT-GAN	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	47.322 ms	2 - 506 MB	NPU
AOT-GAN	TFLITE	float	Snapdragon® 8 Elite Mobile	69.428 ms	2 - 596 MB	NPU
AOT-GAN	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	88.396 ms	4 - 719 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® QCS8275 (Proxy)	541.208 ms	3 - 559 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® QCS8550 (Proxy)	119.169 ms	4 - 7 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA8775P	161.741 ms	3 - 558 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA8775P	161.741 ms	3 - 558 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA8775P	161.741 ms	3 - 558 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® QCS9075	211.015 ms	2 - 45 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® QCS8450 (Proxy)	198.651 ms	3 - 634 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA7255P	541.208 ms	3 - 559 MB	NPU
AOT-GAN	TFLITE	float	Qualcomm® SA8295P	178.621 ms	3 - 493 MB	NPU
AOT-GAN	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	69.428 ms	2 - 596 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.