Yolo-R: Optimized for Qualcomm Devices

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

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

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Input resolution: 640x640
• Model checkpoint: yolor_p6
• Model size (float): 17.9 MB
• Number of parameters: 4.68M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Yolo-R	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	14.486 ms	1 - 328 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	13.576 ms	1 - 320 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® X2 Elite	15.19 ms	2 - 2 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® X Elite	23.229 ms	37 - 37 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	15.874 ms	3 - 393 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	30.311 ms	0 - 389 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	21.484 ms	1 - 7 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	22.502 ms	0 - 40 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® QCS8450	30.311 ms	0 - 389 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	23.232 ms	1 - 6 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	23.229 ms	37 - 37 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	34.113 ms	3 - 307 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	13.576 ms	1 - 320 MB	NPU
Yolo-R	ONNX

License

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

References

• You Only Learn One Representation: Unified Network for Multiple Tasks
• 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.