YOLOv8-Segmentation: Optimized for Qualcomm Devices
Ultralytics YOLOv8 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 YOLOv8-Segmentation 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 YOLOv8-Segmentation on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.semantic_segmentation
Model Stats:
- Model checkpoint: YOLOv8N-Seg
- Input resolution: 640x640
- Number of output classes: 80
- Number of parameters: 3.43M
- Model size (float): 13.2 MB
- Model size (w8a16): 3.91 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| YOLOv8-Segmentation | ONNX | float | Snapdragon® X2 Elite | 3.611 ms | 208 - 208 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Snapdragon® X Elite | 7.04 ms | 144 - 144 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 5.959 ms | 17 - 233 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 12.449 ms | 17 - 213 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Qualcomm® QCS8550 (Proxy) | 7.425 ms | 12 - 25 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Qualcomm® QCS8450 | 12.449 ms | 17 - 213 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Snapdragon® 8 Elite Mobile | 4.66 ms | 13 - 205 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.167 ms | 1 - 214 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Qualcomm® QCS9075 | 8.593 ms | 15 - 60 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Qualcomm® QCS8750 | 4.66 ms | 13 - 205 MB | NPU |
| YOLOv8-Segmentation | ONNX | float | Qualcomm® QCS7181 | 7.04 ms | 144 - 144 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Snapdragon® X2 Elite | 2.862 ms | 5 - 5 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Snapdragon® X Elite | 5.126 ms | 5 - 5 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 3.434 ms | 5 - 210 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 9.353 ms | 4 - 198 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Qualcomm® QCS8275 | 17.1 ms | 1 - 180 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 4.584 ms | 5 - 6 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Qualcomm® QCS8450 | 9.353 ms | 4 - 198 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 2.844 ms | 5 - 188 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Qualcomm® SA8295P | 9.416 ms | 0 - 166 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.003 ms | 5 - 200 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Qualcomm® SA7255P | 17.1 ms | 1 - 180 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Qualcomm® QCS9075 | 6.064 ms | 7 - 17 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Qualcomm® QCS8750 | 2.844 ms | 5 - 188 MB | NPU |
| YOLOv8-Segmentation | QNN_DLC | float | Qualcomm® QCS7181 | 5.126 ms | 5 - 5 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 3.028 ms | 0 - 110 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 8.587 ms | 0 - 84 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® QCS8275 | 16.292 ms | 4 - 81 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 4.114 ms | 4 - 7 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® SA8775P | 34.199 ms | 7 - 39 MB | GPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® SA8650P | 34.199 ms | 7 - 39 MB | GPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® SA8255P | 34.199 ms | 7 - 39 MB | GPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® QCS8450 | 8.587 ms | 0 - 84 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Snapdragon® 8 Elite Mobile | 2.264 ms | 0 - 84 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® SA8295P | 8.636 ms | 4 - 60 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.842 ms | 0 - 94 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® SA7255P | 16.292 ms | 4 - 81 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® QCS9075 | 5.845 ms | 4 - 23 MB | NPU |
| YOLOv8-Segmentation | TFLITE | float | Qualcomm® QCS8750 | 2.264 ms | 0 - 84 MB | NPU |
License
- The license for the original implementation of YOLOv8-Segmentation 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.