Upload README.md with huggingface_hub

53f023a verified 15 days ago

11.9 kB

	---
	library_name: pytorch
	license: agpl-3.0
	pipeline_tag: object-detection
	tags:
	- real_time
	- android

	---

	![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yolov11_det/web-assets/model_demo.png)

	# YOLOv11-Detection: Optimized for Mobile Deployment
	## Real-time object detection optimized for mobile and edge by Ultralytics


	Ultralytics YOLOv11 is a machine learning model that predicts bounding boxes and classes of objects in an image.

	This model is an implementation of YOLOv11-Detection found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect).


	This repository provides scripts to run YOLOv11-Detection on Qualcomm® devices.
	More details on model performance across various devices, can be found
	[here](https://aihub.qualcomm.com/models/yolov11_det).


	### Model Details

	- Model Type: Object detection
	- Model Stats:
	- Model checkpoint: YOLO11-N
	- Input resolution: 640x640
	- Number of parameters: None
	- Model size: None

	\| Model \| Device \| Chipset \| Target Runtime \| Inference Time (ms) \| Peak Memory Range (MB) \| Precision \| Primary Compute Unit \| Target Model
	\|---\|---\|---\|---\|---\|---\|---\|---\|---\|
	\| YOLOv11-Detection \| Samsung Galaxy S23 \| Snapdragon® 8 Gen 2 \| TFLITE \| 5.632 ms \| 0 - 15 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| Samsung Galaxy S23 \| Snapdragon® 8 Gen 2 \| QNN \| 5.581 ms \| 5 - 21 MB \| FP16 \| NPU \| [YOLOv11-Detection.so](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.so) \|
	\| YOLOv11-Detection \| Samsung Galaxy S23 \| Snapdragon® 8 Gen 2 \| ONNX \| 6.047 ms \| 5 - 10 MB \| FP16 \| NPU \| [YOLOv11-Detection.onnx](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.onnx) \|
	\| YOLOv11-Detection \| Samsung Galaxy S24 \| Snapdragon® 8 Gen 3 \| TFLITE \| 3.87 ms \| 0 - 43 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| Samsung Galaxy S24 \| Snapdragon® 8 Gen 3 \| QNN \| 3.886 ms \| 5 - 58 MB \| FP16 \| NPU \| [YOLOv11-Detection.so](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.so) \|
	\| YOLOv11-Detection \| Samsung Galaxy S24 \| Snapdragon® 8 Gen 3 \| ONNX \| 4.159 ms \| 5 - 124 MB \| FP16 \| NPU \| [YOLOv11-Detection.onnx](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.onnx) \|
	\| YOLOv11-Detection \| Snapdragon 8 Elite QRD \| Snapdragon® 8 Elite \| TFLITE \| 3.822 ms \| 0 - 43 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| Snapdragon 8 Elite QRD \| Snapdragon® 8 Elite \| QNN \| 3.861 ms \| 5 - 52 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| Snapdragon 8 Elite QRD \| Snapdragon® 8 Elite \| ONNX \| 3.989 ms \| 5 - 81 MB \| FP16 \| NPU \| [YOLOv11-Detection.onnx](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.onnx) \|
	\| YOLOv11-Detection \| QCS8550 (Proxy) \| QCS8550 Proxy \| TFLITE \| 5.452 ms \| 0 - 15 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| QCS8550 (Proxy) \| QCS8550 Proxy \| QNN \| 5.366 ms \| 5 - 6 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| SA7255P ADP \| SA7255P \| TFLITE \| 59.989 ms \| 0 - 38 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| SA7255P ADP \| SA7255P \| QNN \| 58.912 ms \| 1 - 11 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| SA8255 (Proxy) \| SA8255P Proxy \| TFLITE \| 5.519 ms \| 0 - 19 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| SA8255 (Proxy) \| SA8255P Proxy \| QNN \| 5.423 ms \| 7 - 8 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| SA8295P ADP \| SA8295P \| TFLITE \| 10.056 ms \| 0 - 33 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| SA8295P ADP \| SA8295P \| QNN \| 9.461 ms \| 1 - 6 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| SA8650 (Proxy) \| SA8650P Proxy \| TFLITE \| 5.539 ms \| 0 - 15 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| SA8650 (Proxy) \| SA8650P Proxy \| QNN \| 5.468 ms \| 5 - 6 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| SA8775P ADP \| SA8775P \| TFLITE \| 8.151 ms \| 0 - 40 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| SA8775P ADP \| SA8775P \| QNN \| 8.055 ms \| 0 - 6 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| QCS8450 (Proxy) \| QCS8450 Proxy \| TFLITE \| 8.668 ms \| 0 - 36 MB \| FP16 \| NPU \| [YOLOv11-Detection.tflite](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.tflite) \|
	\| YOLOv11-Detection \| QCS8450 (Proxy) \| QCS8450 Proxy \| QNN \| 8.476 ms \| 5 - 45 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| Snapdragon X Elite CRD \| Snapdragon® X Elite \| QNN \| 5.873 ms \| 5 - 5 MB \| FP16 \| NPU \| Use Export Script \|
	\| YOLOv11-Detection \| Snapdragon X Elite CRD \| Snapdragon® X Elite \| ONNX \| 6.823 ms \| 5 - 5 MB \| FP16 \| NPU \| [YOLOv11-Detection.onnx](https://huggingface.co/qualcomm/YOLOv11-Detection/blob/main/YOLOv11-Detection.onnx) \|




	## Installation

	This model can be installed as a Python package via pip.

	```bash
	pip install "qai-hub-models[yolov11_det]"
	```



	## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device

	Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
	Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.

	With this API token, you can configure your client to run models on the cloud
	hosted devices.
	```bash
	qai-hub configure --api_token API_TOKEN
	```
	Navigate to [docs](https://app.aihub.qualcomm.com/docs/) for more information.



	## Demo off target

	The package contains a simple end-to-end demo that downloads pre-trained
	weights and runs this model on a sample input.

	```bash
	python -m qai_hub_models.models.yolov11_det.demo
	```

	The above demo runs a reference implementation of pre-processing, model
	inference, and post processing.

	NOTE: If you want running in a Jupyter Notebook or Google Colab like
	environment, please add the following to your cell (instead of the above).
	```
	%run -m qai_hub_models.models.yolov11_det.demo
	```


	### Run model on a cloud-hosted device

	In addition to the demo, you can also run the model on a cloud-hosted Qualcomm®
	device. This script does the following:
	* Performance check on-device on a cloud-hosted device
	* Downloads compiled assets that can be deployed on-device for Android.
	* Accuracy check between PyTorch and on-device outputs.

	```bash
	python -m qai_hub_models.models.yolov11_det.export
	```
	```
	Profiling Results
	------------------------------------------------------------
	YOLOv11-Detection
	Device : Samsung Galaxy S23 (13)
	Runtime : TFLITE
	Estimated inference time (ms) : 5.6
	Estimated peak memory usage (MB): [0, 15]
	Total # Ops : 382
	Compute Unit(s) : NPU (382 ops)
	```


	## How does this work?

	This [export script](https://aihub.qualcomm.com/models/yolov11_det/qai_hub_models/models/YOLOv11-Detection/export.py)
	leverages [Qualcomm® AI Hub](https://aihub.qualcomm.com/) to optimize, validate, and deploy this model
	on-device. Lets go through each step below in detail:

	Step 1: Compile model for on-device deployment

	To compile a PyTorch model for on-device deployment, we first trace the model
	in memory using the `jit.trace` and then call the `submit_compile_job` API.

	```python
	import torch

	import qai_hub as hub
	from qai_hub_models.models.yolov11_det import Model

	# Load the model
	torch_model = Model.from_pretrained()

	# Device
	device = hub.Device("Samsung Galaxy S23")

	# Trace model
	input_shape = torch_model.get_input_spec()
	sample_inputs = torch_model.sample_inputs()

	pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

	# Compile model on a specific device
	compile_job = hub.submit_compile_job(
	model=pt_model,
	device=device,
	input_specs=torch_model.get_input_spec(),
	)

	# Get target model to run on-device
	target_model = compile_job.get_target_model()

	```


	Step 2: Performance profiling on cloud-hosted device

	After compiling models from step 1. Models can be profiled model on-device using the
	`target_model`. Note that this scripts runs the model on a device automatically
	provisioned in the cloud. Once the job is submitted, you can navigate to a
	provided job URL to view a variety of on-device performance metrics.
	```python
	profile_job = hub.submit_profile_job(
	model=target_model,
	device=device,
	)

	```

	Step 3: Verify on-device accuracy

	To verify the accuracy of the model on-device, you can run on-device inference
	on sample input data on the same cloud hosted device.
	```python
	input_data = torch_model.sample_inputs()
	inference_job = hub.submit_inference_job(
	model=target_model,
	device=device,
	inputs=input_data,
	)
	on_device_output = inference_job.download_output_data()

	```
	With the output of the model, you can compute like PSNR, relative errors or
	spot check the output with expected output.

	Note: This on-device profiling and inference requires access to Qualcomm®
	AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).



	## Run demo on a cloud-hosted device

	You can also run the demo on-device.

	```bash
	python -m qai_hub_models.models.yolov11_det.demo --on-device
	```

	NOTE: If you want running in a Jupyter Notebook or Google Colab like
	environment, please add the following to your cell (instead of the above).
	```
	%run -m qai_hub_models.models.yolov11_det.demo -- --on-device
	```


	## Deploying compiled model to Android


	The models can be deployed using multiple runtimes:
	- TensorFlow Lite (`.tflite` export): [This
	tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a
	guide to deploy the .tflite model in an Android application.


	- QNN (`.so` export ): This [sample
	app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)
	provides instructions on how to use the `.so` shared library in an Android application.


	## View on Qualcomm® AI Hub
	Get more details on YOLOv11-Detection's performance across various devices [here](https://aihub.qualcomm.com/models/yolov11_det).
	Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)


	## License
	* The license for the original implementation of YOLOv11-Detection can be found [here](https://github.com/ultralytics/ultralytics/blob/main/LICENSE).
	* The license for the compiled assets for on-device deployment can be found [here](https://github.com/ultralytics/ultralytics/blob/main/LICENSE)



	## References
	* [Ultralytics YOLOv11 Docs: Object Detection](https://docs.ultralytics.com/tasks/detect/)
	* [Source Model Implementation](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect)



	## Community
	* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
	* For questions or feedback please [reach out to us](mailto:[email protected]).