--- library_name: pytorch license: agpl-3.0 tags: - real_time - android pipeline_tag: image-segmentation --- ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yolov8_seg/web-assets/model_demo.png) # 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](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). ### Model Details - **Model Type:** Semantic segmentation - **Model Stats:** - Model checkpoint: YOLOv8N-Seg - Input resolution: 640x640 - Number of parameters: 3.43M - Model size: 13.2 MB - Number of output classes: 80 | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model |---|---|---|---|---|---|---|---|---| | YOLOv8-Segmentation | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 8.23 ms | 4 - 26 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 6.154 ms | 5 - 22 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 6.466 ms | 12 - 47 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 5.993 ms | 3 - 53 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 4.31 ms | 51 - 106 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 4.408 ms | 16 - 82 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 4.728 ms | 4 - 47 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 3.483 ms | 5 - 51 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 3.573 ms | 4 - 55 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA7255P ADP | SA7255P | TFLITE | 94.179 ms | 4 - 45 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA7255P ADP | SA7255P | QNN | 90.395 ms | 1 - 10 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 8.121 ms | 4 - 22 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA8255 (Proxy) | SA8255P Proxy | QNN | 4.978 ms | 5 - 7 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA8295P ADP | SA8295P | TFLITE | 12.528 ms | 4 - 33 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA8295P ADP | SA8295P | QNN | 9.108 ms | 0 - 18 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 8.094 ms | 4 - 24 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA8650 (Proxy) | SA8650P Proxy | QNN | 5.076 ms | 5 - 8 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA8775P ADP | SA8775P | TFLITE | 11.909 ms | 4 - 45 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | SA8775P ADP | SA8775P | QNN | 8.095 ms | 1 - 11 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 94.179 ms | 4 - 45 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 90.395 ms | 1 - 10 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 8.248 ms | 5 - 26 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 4.914 ms | 5 - 8 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 11.909 ms | 4 - 45 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 8.095 ms | 1 - 11 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 11.636 ms | 4 - 44 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 9.636 ms | 5 - 39 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 5.444 ms | 5 - 5 MB | FP16 | NPU | -- | | YOLOv8-Segmentation | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.793 ms | 17 - 17 MB | FP16 | NPU | -- | ## License * The license for the original implementation of YOLOv8-Segmentation can be found [here](https://github.com/ultralytics/ultralytics/blob/main/LICENSE). * The license for the compiled assets for on-device deployment can be found [here](https://github.com/ultralytics/ultralytics/blob/main/LICENSE) ## References * [Ultralytics YOLOv8 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://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. * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com). ## 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