--- library_name: pytorch license: other tags: - real_time - android pipeline_tag: object-detection --- ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/rtmdet/web-assets/model_demo.png) # RTMDet: Optimized for Mobile Deployment ## Real-time object detection optimized for mobile and edge 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 model is an implementation of RTMDet found [here](https://github.com/open-mmlab/mmdetection/tree/3.x/configs/rtmdet). This repository provides scripts to run RTMDet on Qualcomm® devices. More details on model performance across various devices, can be found [here](https://aihub.qualcomm.com/models/rtmdet). **WARNING**: The model assets are not readily available for download due to licensing restrictions. ### Model Details - **Model Type:** Model_use_case.object_detection - **Model Stats:** - Model checkpoint: RTMDet Medium - Input resolution: 640x640 - Number of parameters: 27.5M - Model size (float): 105 MB | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model |---|---|---|---|---|---|---|---|---| | RTMDet | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 85.371 ms | 0 - 66 MB | NPU | -- | | RTMDet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 32.048 ms | 0 - 115 MB | NPU | -- | | RTMDet | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 16.164 ms | 0 - 13 MB | NPU | -- | | RTMDet | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 23.717 ms | 0 - 66 MB | NPU | -- | | RTMDet | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 85.371 ms | 0 - 66 MB | NPU | -- | | RTMDet | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 16.164 ms | 0 - 12 MB | NPU | -- | | RTMDet | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 34.815 ms | 0 - 75 MB | NPU | -- | | RTMDet | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 16.185 ms | 0 - 14 MB | NPU | -- | | RTMDet | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 23.717 ms | 0 - 66 MB | NPU | -- | | RTMDet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 15.883 ms | 0 - 17 MB | NPU | -- | | RTMDet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 14.508 ms | 5 - 17 MB | NPU | -- | | RTMDet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 12.319 ms | 0 - 106 MB | NPU | -- | | RTMDet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 11.169 ms | 0 - 45 MB | NPU | -- | | RTMDet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 11.317 ms | 0 - 70 MB | NPU | -- | | RTMDet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 10.348 ms | 1 - 41 MB | NPU | -- | | RTMDet | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 15.985 ms | 51 - 51 MB | NPU | -- | ## Installation Install the package via pip: ```bash pip install "qai-hub-models[rtmdet]" torch==2.4.1 --trusted-host download.openmmlab.com -f https://download.openmmlab.com/mmcv/dist/cpu/torch2.4/index.html -f https://qaihub-public-python-wheels.s3.us-west-2.amazonaws.com/index.html ``` ## 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.rtmdet.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.rtmdet.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.rtmdet.export ``` ## How does this work? This [export script](https://aihub.qualcomm.com/models/rtmdet/qai_hub_models/models/RTMDet/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.rtmdet import Model # Load the model torch_model = Model.from_pretrained() # Device device = hub.Device("Samsung Galaxy S24") # 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.rtmdet.demo --eval-mode 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.rtmdet.demo -- --eval-mode 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 RTMDet's performance across various devices [here](https://aihub.qualcomm.com/models/rtmdet). Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/) ## License * The license for the original implementation of RTMDet can be found [here](https://github.com/open-mmlab/mmdetection/blob/3.x/LICENSE). * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf) ## References * [RTMDet: An Empirical Study of Designing Real-Time Object Detectors](https://github.com/open-mmlab/mmdetection/blob/3.x/README.md) * [Source Model Implementation](https://github.com/open-mmlab/mmdetection/tree/3.x/configs/rtmdet) ## 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:ai-hub-support@qti.qualcomm.com).