ResNet-Mixed-Convolution: Optimized for Qualcomm Devices

ResNet Mixed Convolutions is a network with a mixture of 2D and 3D convolutions used for video understanding.

This is based on the implementation of ResNet-Mixed-Convolution 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
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a16	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

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

Model Details

Model Type: Model_use_case.video_classification

Model Stats:

• Model checkpoint: Kinetics-400
• Input resolution: 112x112
• Number of parameters: 11.7M
• Model size (float): 44.6 MB
• Model size (w8a16): 11.5 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ResNet-Mixed-Convolution	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	5.557 ms	2 - 221 MB	NPU
ResNet-Mixed-Convolution	ONNX	float	Snapdragon® 8 Elite Mobile	7.828 ms	0 - 215 MB	NPU
ResNet-Mixed-Convolution	ONNX	float	Snapdragon® X2 Elite	7.097 ms	22 - 22 MB	NPU
ResNet-Mixed-Convolution	ONNX	float	Snapdragon® X Elite	13.846 ms	22 - 22 MB	NPU
ResNet-Mixed-Convolution	ONNX	float	Snapdragon® X Elite	13.846 ms	22 - 22 MB	NPU
ResNet-Mixed-Convolution	ONNX	float	Snapdragon® 8 Gen 3 Mobile	9.621 ms	2 - 275 MB	NPU
ResNet-Mixed-Convolution	ONNX	float	Qualcomm® QCS8550 (Proxy)	13.176 ms	0 - 28 MB	NPU
ResNet-Mixed-Convolution	ONNX	float	Qualcomm® QCS9075	26.23 ms	2 - 5 MB	NPU
ResNet-Mixed-Convolution	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	7.828 ms	0 - 215 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	3.513 ms	0 - 191 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® 8 Elite Mobile	5.245 ms	1 - 192 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® X2 Elite	4.644 ms	12 - 12 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® X Elite	9.055 ms	11 - 11 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® X Elite	9.055 ms	11 - 11 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	6.39 ms	1 - 257 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Qualcomm® QCS6490	1727.557 ms	44 - 62 MB	CPU
ResNet-Mixed-Convolution	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	8.531 ms	1 - 5 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Qualcomm® QCS9075	9.035 ms	1 - 4 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Qualcomm® QCM6690	895.757 ms	106 - 113 MB	CPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	5.245 ms	1 - 192 MB	NPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	878.876 ms	101 - 109 MB	CPU
ResNet-Mixed-Convolution	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	878.876 ms	101 - 109 MB	CPU
ResNet-Mixed-Convolution	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	5.767 ms	2 - 232 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Snapdragon® 8 Elite Mobile	7.64 ms	0 - 227 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Snapdragon® X2 Elite	7.728 ms	2 - 2 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Snapdragon® X Elite	14.124 ms	2 - 2 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Snapdragon® X Elite	14.124 ms	2 - 2 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	9.523 ms	2 - 278 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	97.155 ms	1 - 223 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	13.381 ms	2 - 27 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® SA8775P	25.202 ms	2 - 224 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® SA8775P	25.202 ms	2 - 224 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® SA8775P	25.202 ms	2 - 224 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® QCS9075	27.801 ms	2 - 6 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	27.657 ms	0 - 238 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® SA7255P	97.155 ms	1 - 223 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Qualcomm® SA8295P	26.868 ms	0 - 195 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	7.64 ms	0 - 227 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	4.033 ms	1 - 195 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	5.347 ms	1 - 187 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® X2 Elite	5.431 ms	1 - 1 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® X Elite	10.072 ms	1 - 1 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® X Elite	10.072 ms	1 - 1 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	6.747 ms	0 - 241 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® QCS6490	37.244 ms	1 - 4 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	30.019 ms	1 - 189 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	9.416 ms	1 - 4 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® SA8775P	9.368 ms	1 - 191 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® SA8775P	9.368 ms	1 - 191 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® SA8775P	9.368 ms	1 - 191 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® QCS9075	10.374 ms	1 - 4 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® QCM6690	172.128 ms	1 - 210 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® QCS8450 (Proxy)	13.021 ms	0 - 244 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® SA7255P	30.019 ms	1 - 189 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Qualcomm® SA8295P	16.13 ms	0 - 193 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	5.347 ms	1 - 187 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	15.935 ms	1 - 199 MB	NPU
ResNet-Mixed-Convolution	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	15.935 ms	1 - 199 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	197.199 ms	0 - 257 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Snapdragon® 8 Elite Mobile	202.848 ms	0 - 257 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	221.931 ms	2 - 305 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® QCS8275 (Proxy)	596.879 ms	0 - 253 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® QCS8550 (Proxy)	308.873 ms	0 - 3 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® SA8775P	312.014 ms	0 - 250 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® SA8775P	312.014 ms	0 - 250 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® SA8775P	312.014 ms	0 - 250 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® QCS9075	314.1 ms	0 - 28 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® QCS8450 (Proxy)	345.601 ms	0 - 270 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® SA7255P	596.879 ms	0 - 253 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Qualcomm® SA8295P	358.731 ms	0 - 234 MB	NPU
ResNet-Mixed-Convolution	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	202.848 ms	0 - 257 MB	NPU

License

• The license for the original implementation of ResNet-Mixed-Convolution can be found here.

References

• A Closer Look at Spatiotemporal Convolutions for Action Recognition
• 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.