YOLOv8-Segmentation: Optimized for Mobile Deployment
Real-time object segmentation optimized for mobile and edge by Ultralytics
Ultralytics YOLOv8 is a machine learning model that predicts bounding boxes, segmentation masks and classes of objects in an image.
This model is an implementation of YOLOv8-Segmentation found here.
More details on model performance across various devices, can be found here.
Model Details
- Model Type: Model_use_case.semantic_segmentation
- Model Stats:
- Model checkpoint: YOLOv8N-Seg
- Input resolution: 640x640
- Number of parameters: 3.43M
- Number of output classes: 80
- Model size (float): 13.2 MB
| 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 | 21.595 ms | 4 - 45 MB | NPU | -- |
| YOLOv8-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 17.644 ms | 3 - 13 MB | NPU | -- |
| YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 11.836 ms | 4 - 42 MB | NPU | -- |
| YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 11.091 ms | 5 - 44 MB | NPU | -- |
| YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 8.05 ms | 4 - 24 MB | NPU | -- |
| YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 4.882 ms | 5 - 8 MB | NPU | -- |
| YOLOv8-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 6.719 ms | 2 - 16 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 21.595 ms | 4 - 45 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 17.644 ms | 3 - 13 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 8.177 ms | 4 - 22 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 4.881 ms | 5 - 7 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 12.502 ms | 4 - 31 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 9.114 ms | 0 - 18 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 8.015 ms | 4 - 24 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 4.867 ms | 5 - 7 MB | NPU | -- |
| YOLOv8-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 6.719 ms | 2 - 16 MB | NPU | -- |
| YOLOv8-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 8.237 ms | 4 - 21 MB | NPU | -- |
| YOLOv8-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 4.851 ms | 5 - 23 MB | NPU | -- |
| YOLOv8-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 6.425 ms | 14 - 49 MB | NPU | -- |
| YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.936 ms | 0 - 52 MB | NPU | -- |
| YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 3.591 ms | 4 - 62 MB | NPU | -- |
| YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 4.432 ms | 15 - 81 MB | NPU | -- |
| YOLOv8-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 4.797 ms | 4 - 49 MB | NPU | -- |
| YOLOv8-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 3.3 ms | 5 - 56 MB | NPU | -- |
| YOLOv8-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 3.66 ms | 5 - 57 MB | NPU | -- |
| YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 5.356 ms | 5 - 5 MB | NPU | -- |
| YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 7.294 ms | 17 - 17 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 9.004 ms | 1 - 11 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 6.582 ms | 2 - 44 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 4.673 ms | 2 - 6 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 5.279 ms | 1 - 16 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 21.094 ms | 2 - 14 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN | 9.004 ms | 1 - 11 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 4.695 ms | 4 - 6 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN | 5.989 ms | 0 - 18 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 4.653 ms | 2 - 5 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN | 5.279 ms | 1 - 16 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 4.663 ms | 2 - 16 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 7.503 ms | 6 - 31 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 3.04 ms | 2 - 46 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 5.281 ms | 9 - 65 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 3.88 ms | 1 - 49 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 5.118 ms | 2 - 2 MB | NPU | -- |
| YOLOv8-Segmentation | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 8.425 ms | 15 - 15 MB | NPU | -- |
License
- The license for the original implementation of YOLOv8-Segmentation can be found here.
- The license for the compiled assets for on-device deployment 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.
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
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support