qualcomm
/

YOLOv11-Segmentation

@@ -10,236 +10,65 @@ pipeline_tag: image-segmentation
 ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yolov11_seg/web-assets/model_demo.png)
-# YOLOv11-Segmentation: Optimized for Mobile Deployment
-## Real-time object segmentation optimized for mobile and edge by Ultralytics
 Ultralytics YOLOv11 is a machine learning model that predicts bounding boxes, segmentation masks and classes of objects in an image.
-This model is an implementation of YOLOv11-Segmentation found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/segment).
-This repository provides scripts to run YOLOv11-Segmentation on Qualcomm® devices.
-More details on model performance across various devices, can be found
-[here](https://aihub.qualcomm.com/models/yolov11_seg).
-**WARNING**: The model assets are not readily available for download due to licensing restrictions.
-### Model Details
-- **Model Type:** Model_use_case.semantic_segmentation
-- **Model Stats:**
-  - Model checkpoint: YOLO11N-Seg
-  - Input resolution: 640x640
-  - Number of output classes: 80
-  - Number of parameters: 2.89M
-  - Model size (float): 11.1 MB
-  - Model size (w8a16): 11.4 MB
-| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
-|---|---|---|---|---|---|---|---|---|
-| YOLOv11-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 15.485 ms | 4 - 224 MB | NPU | -- |
-| YOLOv11-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 10.778 ms | 4 - 187 MB | NPU | -- |
-| YOLOv11-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 4.353 ms | 4 - 7 MB | NPU | -- |
-| YOLOv11-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 6.335 ms | 14 - 19 MB | NPU | -- |
-| YOLOv11-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 6.157 ms | 4 - 205 MB | NPU | -- |
-| YOLOv11-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 15.485 ms | 4 - 224 MB | NPU | -- |
-| YOLOv11-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 9.355 ms | 4 - 169 MB | NPU | -- |
-| YOLOv11-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 6.157 ms | 4 - 205 MB | NPU | -- |
-| YOLOv11-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 3.2 ms | 0 - 375 MB | NPU | -- |
-| YOLOv11-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 4.125 ms | 15 - 227 MB | NPU | -- |
-| YOLOv11-Segmentation | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 2.414 ms | 0 - 216 MB | NPU | -- |
-| YOLOv11-Segmentation | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 3.971 ms | 2 - 182 MB | NPU | -- |
-| YOLOv11-Segmentation | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 1.968 ms | 0 - 236 MB | NPU | -- |
-| YOLOv11-Segmentation | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 3.117 ms | 1 - 153 MB | NPU | -- |
-| YOLOv11-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.373 ms | 17 - 17 MB | NPU | -- |
-## Installation
-Install the package via pip:
-```bash
-# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
-pip install "qai-hub-models[yolov11-seg]"
-```
-## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
-Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.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://workbench.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_seg.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_seg.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_seg.export
-```
-## How does this work?
-This [export script](https://aihub.qualcomm.com/models/yolov11_seg/qai_hub_models/models/YOLOv11-Segmentation/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_seg import Model
-# Load the model
-torch_model = Model.from_pretrained()
-# Device
-device = hub.Device("Samsung Galaxy S25")
-# 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 Workbench. [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_seg.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.yolov11_seg.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 YOLOv11-Segmentation's performance across various devices [here](https://aihub.qualcomm.com/models/yolov11_seg).
-Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
 ## License
 * The license for the original implementation of YOLOv11-Segmentation can be found
   [here](https://github.com/ultralytics/ultralytics/blob/main/LICENSE).
 ## References
 * [Ultralytics YOLOv11 Docs: Instance Segmentation](https://docs.ultralytics.com/tasks/segment/)
 * [Source Model Implementation](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/segment)
 ## 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).

 ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yolov11_seg/web-assets/model_demo.png)
+# YOLOv11-Segmentation: Optimized for Qualcomm Devices
 Ultralytics YOLOv11 is a machine learning model that predicts bounding boxes, segmentation masks and classes of objects in an image.
+This is based on the implementation of YOLOv11-Segmentation found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/segment).
+This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/yolov11_seg) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
+Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/yolov11_seg) 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 [YOLOv11-Segmentation on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/yolov11_seg) for usage instructions.
+## Model Details
+**Model Type:** Model_use_case.semantic_segmentation
+**Model Stats:**
+- Model checkpoint: YOLO11N-Seg
+- Input resolution: 640x640
+- Number of output classes: 80
+- Number of parameters: 2.89M
+- Model size (float): 11.1 MB
+- Model size (w8a16): 11.4 MB
+## Performance Summary
+| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
+|---|---|---|---|---|---|---
+| YOLOv11-Segmentation | ONNX | float | Snapdragon® X Elite | 6.416 ms | 17 - 17 MB | NPU
+| YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 4.233 ms | 2 - 206 MB | NPU
+| YOLOv11-Segmentation | ONNX | float | Qualcomm® QCS8550 (Proxy) | 6.171 ms | 11 - 15 MB | NPU
+| YOLOv11-Segmentation | ONNX | float | Qualcomm® QCS9075 | 8.053 ms | 11 - 14 MB | NPU
+| YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.491 ms | 1 - 160 MB | NPU
+| YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.043 ms | 0 - 155 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 3.147 ms | 0 - 171 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 15.466 ms | 4 - 111 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 4.342 ms | 4 - 9 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA8775P | 6.082 ms | 4 - 113 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS9075 | 5.829 ms | 4 - 22 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 10.084 ms | 4 - 210 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA7255P | 15.466 ms | 4 - 111 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA8295P | 9.237 ms | 4 - 177 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.39 ms | 0 - 116 MB | NPU
+| YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.957 ms | 0 - 121 MB | NPU
 ## License
 * The license for the original implementation of YOLOv11-Segmentation can be found
   [here](https://github.com/ultralytics/ultralytics/blob/main/LICENSE).
 ## References
 * [Ultralytics YOLOv11 Docs: Instance Segmentation](https://docs.ultralytics.com/tasks/segment/)
 * [Source Model Implementation](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/segment)
 ## 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).