Yolo-v3: Optimized for Qualcomm Devices

YoloV3 is a machine learning model that predicts bounding boxes and classes of objects in an image.

This is based on the implementation of Yolo-v3 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-v3 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Model checkpoint: YoloV3 Tiny
• Input resolution: 416p (416x416)
• Number of parameters: 11.5M
• Model size (float): 43.9 MB
• Model size (w8a16): 16.9 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Yolo-v3	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1.814 ms	0 - 150 MB	NPU
Yolo-v3	ONNX	float	Snapdragon® X2 Elite	2.421 ms	20 - 20 MB	NPU
Yolo-v3	ONNX	float	Snapdragon® X Elite	4.187 ms	19 - 19 MB	NPU
Yolo-v3	ONNX	float	Snapdragon® 8 Gen 3 Mobile	3.235 ms	2 - 171 MB	NPU
Yolo-v3	ONNX	float	Qualcomm® QCS8550 (Proxy)	3.994 ms	0 - 45 MB	NPU
Yolo-v3	ONNX	float	Qualcomm® QCS9075	6.22 ms	5 - 7 MB	NPU
Yolo-v3	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	2.523 ms	3 - 154 MB	NPU
Yolo-v3	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	1.39 ms	0 - 56 MB	NPU
Yolo-v3	ONNX	w8a16	Snapdragon® X2 Elite	1.612 ms	10 - 10 MB	NPU
Yolo-v3	ONNX	w8a16	Snapdragon® X Elite	4.016 ms	10 - 10 MB	NPU
Yolo-v3	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	2.547 ms	0 - 93 MB	NPU
Yolo-v3	ONNX	w8a16	Qualcomm® QCS6490	530.455 ms	213 - 218 MB	CPU
Yolo-v3	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	3.704 ms	0 - 12 MB	NPU
Yolo-v3	ONNX	w8a16	Qualcomm® QCS9075	4.006 ms	2 - 5 MB	NPU
Yolo-v3	ONNX	w8a16	Qualcomm® QCM6690	261.29 ms	216 - 223 MB	CPU
Yolo-v3	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	1.884 ms	0 - 53 MB	NPU
Yolo-v3	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	232.941 ms	215 - 222 MB	CPU
Yolo-v3	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.731 ms	5 - 152 MB	NPU
Yolo-v3	QNN_DLC	float	Snapdragon® X2 Elite	2.421 ms	5 - 5 MB	NPU
Yolo-v3	QNN_DLC	float	Snapdragon® X Elite	3.679 ms	5 - 5 MB	NPU
Yolo-v3	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	2.73 ms	0 - 159 MB	NPU
Yolo-v3	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	20.096 ms	0 - 143 MB	NPU
Yolo-v3	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	3.386 ms	2 - 3 MB	NPU
Yolo-v3	QNN_DLC	float	Qualcomm® QCS9075	5.749 ms	7 - 13 MB	NPU
Yolo-v3	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	10.09 ms	4 - 173 MB	NPU
Yolo-v3	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	2.18 ms	5 - 150 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	1.378 ms	2 - 53 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Snapdragon® X2 Elite	1.815 ms	2 - 2 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Snapdragon® X Elite	4.07 ms	2 - 2 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	2.574 ms	2 - 81 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Qualcomm® QCS6490	11.734 ms	1 - 5 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	9.23 ms	1 - 46 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	3.717 ms	2 - 129 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Qualcomm® QCS9075	3.949 ms	2 - 6 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Qualcomm® QCM6690	38.202 ms	2 - 174 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Qualcomm® QCS8450 (Proxy)	4.368 ms	2 - 79 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	1.898 ms	2 - 55 MB	NPU
Yolo-v3	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	4.716 ms	2 - 174 MB	NPU
Yolo-v3	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.513 ms	0 - 148 MB	NPU
Yolo-v3	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.691 ms	0 - 187 MB	NPU
Yolo-v3	TFLITE	float	Qualcomm® QCS8275 (Proxy)	19.965 ms	0 - 146 MB	NPU
Yolo-v3	TFLITE	float	Qualcomm® QCS8550 (Proxy)	3.308 ms	0 - 1 MB	NPU
Yolo-v3	TFLITE	float	Qualcomm® QCS9075	5.666 ms	0 - 30 MB	NPU
Yolo-v3	TFLITE	float	Qualcomm® QCS8450 (Proxy)	9.893 ms	0 - 192 MB	NPU
Yolo-v3	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	2.132 ms	0 - 148 MB	NPU

License

• The license for the original implementation of Yolo-v3 can be found here.

References

• YOLOv3: An Incremental Improvement
• 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.