Yolo-v5: Optimized for Qualcomm Devices

YoloV5 is a machine learning model that predicts bounding boxes and classes of objects in an image.

This is based on the implementation of Yolo-v5 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

Due to licensing restrictions, we cannot distribute pre-exported model assets for this model. 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

See our repository for Yolo-v5 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Model checkpoint: YoloV5-M
• Input resolution: 640x640
• Number of parameters: 21.2M
• Model size (float): 81.1 MB
• Model size (w8a16): 21.8 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Yolo-v5	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	5.269 ms	5 - 217 MB	NPU
Yolo-v5	ONNX	float	Snapdragon® X2 Elite	7.47 ms	46 - 46 MB	NPU
Yolo-v5	ONNX	float	Snapdragon® X Elite	13.547 ms	46 - 46 MB	NPU
Yolo-v5	ONNX	float	Snapdragon® 8 Gen 3 Mobile	9.672 ms	2 - 282 MB	NPU
Yolo-v5	ONNX	float	Qualcomm® QCS8550 (Proxy)	13.023 ms	0 - 307 MB	NPU
Yolo-v5	ONNX	float	Qualcomm® QCS9075	22.252 ms	5 - 12 MB	NPU
Yolo-v5	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	7.311 ms	1 - 230 MB	NPU
Yolo-v5	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	4.081 ms	3 - 265 MB	NPU
Yolo-v5	ONNX	w8a16	Snapdragon® X2 Elite	4.881 ms	24 - 24 MB	NPU
Yolo-v5	ONNX	w8a16	Snapdragon® X Elite	10.285 ms	23 - 23 MB	NPU
Yolo-v5	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	6.206 ms	2 - 311 MB	NPU
Yolo-v5	ONNX	w8a16	Qualcomm® QCS6490	1752.073 ms	94 - 97 MB	CPU
Yolo-v5	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	9.601 ms	0 - 29 MB	NPU
Yolo-v5	ONNX	w8a16	Qualcomm® QCS9075	12.475 ms	2 - 5 MB	NPU
Yolo-v5	ONNX	w8a16	Qualcomm® QCM6690	871.454 ms	93 - 103 MB	CPU
Yolo-v5	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	5.38 ms	1 - 259 MB	NPU
Yolo-v5	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	834.407 ms	83 - 94 MB	CPU
Yolo-v5	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	4.767 ms	5 - 207 MB	NPU
Yolo-v5	QNN_DLC	float	Snapdragon® X2 Elite	6.682 ms	5 - 5 MB	NPU
Yolo-v5	QNN_DLC	float	Snapdragon® X Elite	11.959 ms	5 - 5 MB	NPU
Yolo-v5	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	8.454 ms	5 - 241 MB	NPU
Yolo-v5	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	63.689 ms	1 - 189 MB	NPU
Yolo-v5	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	11.274 ms	5 - 73 MB	NPU
Yolo-v5	QNN_DLC	float	Qualcomm® QCS9075	18.082 ms	5 - 11 MB	NPU
Yolo-v5	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	25.862 ms	5 - 263 MB	NPU
Yolo-v5	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	6.383 ms	5 - 202 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	3.223 ms	2 - 244 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Snapdragon® X2 Elite	4.301 ms	2 - 2 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Snapdragon® X Elite	9.478 ms	2 - 2 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	5.825 ms	2 - 284 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Qualcomm® QCS6490	32.005 ms	1 - 5 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	21.071 ms	2 - 233 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	8.758 ms	2 - 226 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Qualcomm® QCS9075	10.527 ms	0 - 4 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Qualcomm® QCM6690	124.7 ms	2 - 269 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Qualcomm® QCS8450 (Proxy)	13.778 ms	2 - 286 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	4.287 ms	2 - 240 MB	NPU
Yolo-v5	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	12.656 ms	2 - 249 MB	NPU
Yolo-v5	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	4.44 ms	0 - 235 MB	NPU
Yolo-v5	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	8.195 ms	0 - 291 MB	NPU
Yolo-v5	TFLITE	float	Qualcomm® QCS8275 (Proxy)	63.119 ms	0 - 222 MB	NPU
Yolo-v5	TFLITE	float	Qualcomm® QCS8550 (Proxy)	10.739 ms	0 - 3 MB	NPU
Yolo-v5	TFLITE	float	Qualcomm® QCS9075	17.861 ms	0 - 57 MB	NPU
Yolo-v5	TFLITE	float	Qualcomm® QCS8450 (Proxy)	30.364 ms	0 - 316 MB	NPU
Yolo-v5	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	6.127 ms	0 - 219 MB	NPU

License

• The license for the original implementation of Yolo-v5 can be found here.

References

• 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.