RTMDet: Optimized for Qualcomm Devices

RTMDet is a highly efficient model for real-time object detection,capable of predicting both the bounding boxes and classes of objects within an image.It is highly optimized for real-time applications, making it reliable for industrial and commercial use

This is based on the implementation of RTMDet 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 RTMDet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

Input resolution: 640x640
Model checkpoint: RTMDet Medium
Model size (float): 105 MB
Number of parameters: 27.5M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
RTMDet	ONNX	float	Snapdragon® X2 Elite	9.068 ms	5 - 5 MB	NPU
RTMDet	ONNX	float	Snapdragon® X Elite	16.503 ms	51 - 51 MB	NPU
RTMDet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	12.388 ms	1 - 237 MB	NPU
RTMDet	ONNX	float	Snapdragon® 8 Gen 1 Mobile	39.929 ms	3 - 299 MB	NPU
RTMDet	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	27.056 ms	5 - 13 MB	NPU
RTMDet	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	16.285 ms	0 - 54 MB	NPU
RTMDet	ONNX	float	Qualcomm® QCS8450	39.929 ms	3 - 299 MB	NPU
RTMDet	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	23.723 ms	5 - 12 MB	NPU
RTMDet	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	16.503 ms	51 - 51 MB	NPU
RTMDet	ONNX	float	Qualcomm® Dragonwing™ Q-8750	9.478 ms	3 - 193 MB	NPU
RTMDet	ONNX	float	Snapdragon® 8 Elite Mobile	9.478 ms	3 - 193 MB	NPU
RTMDet	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	7.191 ms	3 - 198 MB	NPU
RTMDet	ONNX	w8a16	Snapdragon® X2 Elite	6.699 ms	2 - 2 MB	NPU
RTMDet	ONNX	w8a16	Snapdragon® X Elite	15.681 ms	27 - 27 MB	NPU
RTMDet	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	9.94 ms	3 - 357 MB	NPU
RTMDet	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	20.392 ms	3 - 357 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	64.692 ms	23 - 26 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	14.03 ms	2 - 6 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	15.177 ms	2 - 5 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® QCS8450	20.392 ms	3 - 357 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	15.575 ms	2 - 5 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	15.681 ms	27 - 27 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	147.35 ms	26 - 346 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	23.259 ms	25 - 353 MB	NPU
RTMDet	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	7.213 ms	0 - 300 MB	NPU
RTMDet	ONNX	w8a16	Snapdragon® 8 Elite Mobile	7.213 ms	0 - 300 MB	NPU
RTMDet	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	5.914 ms	1 - 328 MB	NPU
RTMDet	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	23.259 ms	25 - 353 MB	NPU
RTMDet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	12.113 ms	9 - 298 MB	NPU
RTMDet	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	46.481 ms	10 - 360 MB	NPU
RTMDet	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	27.394 ms	9 - 81 MB	NPU
RTMDet	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	15.89 ms	9 - 13 MB	NPU
RTMDet	TFLITE	float	Qualcomm® SA8775P	24.277 ms	9 - 223 MB	NPU
RTMDet	TFLITE	float	Qualcomm® SA8650P	24.277 ms	9 - 223 MB	NPU
RTMDet	TFLITE	float	Qualcomm® SA8255P	24.277 ms	9 - 223 MB	NPU
RTMDet	TFLITE	float	Qualcomm® QCS8450	46.481 ms	10 - 360 MB	NPU
RTMDet	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	23.87 ms	8 - 79 MB	NPU
RTMDet	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	9.091 ms	9 - 225 MB	NPU
RTMDet	TFLITE	float	Qualcomm® SA7255P	85.551 ms	10 - 221 MB	NPU
RTMDet	TFLITE	float	Qualcomm® SA8295P	33.087 ms	9 - 283 MB	NPU
RTMDet	TFLITE	float	Snapdragon® 8 Elite Mobile	9.091 ms	9 - 225 MB	NPU
RTMDet	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	6.897 ms	8 - 224 MB	NPU

License

The license for the original implementation of RTMDet can be found here.

References

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.

Downloads last month: -; Downloads are not tracked for this model. How to track