v0.34.0

See https://github.com/quic/ai-hub-models/releases/v0.34.0 for changelog.

README.md

@@ -19,7 +19,11 @@ Ultralytics YOLOv8 is a machine learning model that predicts bounding boxes, seg
 This model is an implementation of YOLOv8-Segmentation found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/segment).
 
 
- More details on model performance across various devices, can be found [here](https://aihub.qualcomm.com/models/yolov8_seg).
+This repository provides scripts to run YOLOv8-Segmentation on Qualcomm® devices.
+More details on model performance across various devices, can be found
+[here](https://aihub.qualcomm.com/models/yolov8_seg).
+
+**WARNING**: The model assets are not readily available for download due to licensing restrictions.
 
 ### Model Details
 
@@ -34,55 +38,223 @@ This model is an implementation of YOLOv8-Segmentation found [here](https://gith
 
 | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
 |---|---|---|---|---|---|---|---|---|
-| YOLOv8-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 21.491 ms | 4 - 65 MB | NPU | -- |
+| YOLOv8-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 21.454 ms | 4 - 66 MB | NPU | -- |
 | YOLOv8-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 17.023 ms | 4 - 117 MB | NPU | -- |
-| YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 11.53 ms | 4 - 50 MB | NPU | -- |
+| YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 11.418 ms | 4 - 47 MB | NPU | -- |
 | YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 10.219 ms | 5 - 43 MB | NPU | -- |
-| YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 8.117 ms | 4 - 27 MB | NPU | -- |
+| YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 8.092 ms | 4 - 24 MB | NPU | -- |
 | YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 4.457 ms | 5 - 41 MB | NPU | -- |
-| YOLOv8-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 10.25 ms | 4 - 66 MB | NPU | -- |
+| YOLOv8-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 10.25 ms | 4 - 65 MB | NPU | -- |
 | YOLOv8-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 6.317 ms | 0 - 112 MB | NPU | -- |
-| YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 21.491 ms | 4 - 65 MB | NPU | -- |
+| YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 21.454 ms | 4 - 66 MB | NPU | -- |
 | YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 17.023 ms | 4 - 117 MB | NPU | -- |
-| YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 8.169 ms | 4 - 28 MB | NPU | -- |
+| YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 8.165 ms | 4 - 27 MB | NPU | -- |
 | YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 4.48 ms | 5 - 48 MB | NPU | -- |
-| YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 12.674 ms | 4 - 34 MB | NPU | -- |
+| YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 12.708 ms | 4 - 34 MB | NPU | -- |
 | YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 8.576 ms | 0 - 34 MB | NPU | -- |
-| YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 8.213 ms | 4 - 26 MB | NPU | -- |
+| YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 8.165 ms | 4 - 30 MB | NPU | -- |
 | YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 4.481 ms | 5 - 35 MB | NPU | -- |
-| YOLOv8-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 10.25 ms | 4 - 66 MB | NPU | -- |
+| YOLOv8-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 10.25 ms | 4 - 65 MB | NPU | -- |
 | YOLOv8-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 6.317 ms | 0 - 112 MB | NPU | -- |
-| YOLOv8-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 8.157 ms | 4 - 27 MB | NPU | -- |
+| YOLOv8-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 8.147 ms | 4 - 27 MB | NPU | -- |
 | YOLOv8-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 4.47 ms | 4 - 43 MB | NPU | -- |
-| YOLOv8-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 6.74 ms | 10 - 93 MB | NPU | -- |
-| YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.902 ms | 3 - 68 MB | NPU | -- |
+| YOLOv8-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 6.813 ms | 0 - 75 MB | NPU | -- |
+| YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.862 ms | 4 - 71 MB | NPU | -- |
 | YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 3.28 ms | 5 - 204 MB | NPU | -- |
-| YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 4.457 ms | 16 - 211 MB | NPU | -- |
-| YOLOv8-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 5.555 ms | 3 - 66 MB | NPU | -- |
+| YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 4.791 ms | 15 - 204 MB | NPU | -- |
+| YOLOv8-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 5.542 ms | 3 - 56 MB | NPU | -- |
 | YOLOv8-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 2.965 ms | 5 - 130 MB | NPU | -- |
-| YOLOv8-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 3.693 ms | 1 - 132 MB | NPU | -- |
+| YOLOv8-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 4.338 ms | 5 - 135 MB | NPU | -- |
 | YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 5.029 ms | 5 - 5 MB | NPU | -- |
-| YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 7.144 ms | 17 - 17 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 8.22 ms | 2 - 36 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 5.137 ms | 2 - 46 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 4.079 ms | 2 - 15 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 4.789 ms | 0 - 36 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 16.582 ms | 2 - 40 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 8.22 ms | 2 - 36 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 4.068 ms | 2 - 15 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 5.446 ms | 1 - 37 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 4.073 ms | 2 - 16 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 4.789 ms | 0 - 36 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 4.08 ms | 2 - 16 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 7.985 ms | 6 - 29 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 2.695 ms | 2 - 45 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 5.164 ms | 9 - 79 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 2.318 ms | 2 - 43 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 5.008 ms | 5 - 103 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 5.021 ms | 0 - 0 MB | NPU | -- |
-| YOLOv8-Segmentation | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 8.782 ms | 15 - 15 MB | NPU | -- |
+| YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 7.18 ms | 17 - 17 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 8.154 ms | 2 - 36 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 5.254 ms | 2 - 40 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 4.067 ms | 2 - 15 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 4.76 ms | 1 - 37 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 15.523 ms | 2 - 39 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 8.154 ms | 2 - 36 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 4.063 ms | 2 - 14 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 5.395 ms | 2 - 40 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 4.071 ms | 2 - 16 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 4.76 ms | 1 - 37 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 4.072 ms | 2 - 16 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 7.953 ms | 2 - 31 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 2.693 ms | 2 - 47 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 5.201 ms | 9 - 76 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 1.917 ms | 2 - 44 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 4.761 ms | 10 - 106 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 4.918 ms | 0 - 0 MB | NPU | -- |
+| YOLOv8-Segmentation | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 8.742 ms | 14 - 14 MB | NPU | -- |
+
+
+
+
+## Installation
+
+
+Install the package via pip:
+```bash
+pip install "qai-hub-models[yolov8-seg]"
+```
+
+
+## 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.yolov8_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.yolov8_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.yolov8_seg.export
+```
+
+
+
+## How does this work?
+
+This [export script](https://aihub.qualcomm.com/models/yolov8_seg/qai_hub_models/models/YOLOv8-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.yolov8_seg 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.yolov8_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.yolov8_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 YOLOv8-Segmentation's performance across various devices [here](https://aihub.qualcomm.com/models/yolov8_seg).
+Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
 
 
 ## License
@@ -99,26 +271,7 @@ This model is an implementation of YOLOv8-Segmentation found [here](https://gith
 
 
 ## Community
-* Join [our AI Hub Slack community](https://qualcomm-ai-hub.slack.com/join/shared_invite/zt-2d5zsmas3-Sj0Q9TzslueCjS31eXG2UA#/shared-invite/email) to collaborate, post questions and learn more about on-device AI.
+* 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]).
 
-## Usage and Limitations
-
-Model may not be used for or in connection with any of the following applications:
-
-- Accessing essential private and public services and benefits;
-- Administration of justice and democratic processes;
-- Assessing or recognizing the emotional state of a person;
-- Biometric and biometrics-based systems, including categorization of persons based on sensitive characteristics;
-- Education and vocational training;
-- Employment and workers management;
-- Exploitation of the vulnerabilities of persons resulting in harmful behavior;
-- General purpose social scoring;
-- Law enforcement;
-- Management and operation of critical infrastructure;
-- Migration, asylum and border control management;
-- Predictive policing;
-- Real-time remote biometric identification in public spaces;
-- Recommender systems of social media platforms;
-- Scraping of facial images (from the internet or otherwise); and/or
-- Subliminal manipulation
+