File size: 6,863 Bytes
f9ead1e 7fed5fa f9ead1e 7fed5fa f9ead1e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 |
---
library_name: pytorch
license: agpl-3.0
tags:
- real_time
- quantized
- android
pipeline_tag: object-detection
---
# YOLOv11-Detection-Quantized: Optimized for Mobile Deployment
## Quantized real-time object detection optimized for mobile and edge by Ultralytics
Ultralytics YOLOv11 is a machine learning model that predicts bounding boxes and classes of objects in an image. This model is post-training quantized to int8 using samples from the COCO dataset.
This model is an implementation of YOLOv11-Detection-Quantized found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect).
More details on model performance across various devices, can be found [here](https://aihub.qualcomm.com/models/yolov11_det_quantized).
### Model Details
- **Model Type:** Object detection
- **Model Stats:**
- Model checkpoint: YOLOv11-N
- Input resolution: 640x640
- Number of parameters: 2.64M
- Model size: 2.83 MB
- Precision: w8a8 (8-bit weights, 8-bit activations)
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|---|---|---|---|---|---|---|---|---|
| YOLOv11-Detection-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.825 ms | 0 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 1.846 ms | 1 - 3 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 10.611 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.208 ms | 0 - 33 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.202 ms | 1 - 20 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 7.764 ms | 1 - 66 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.92 ms | 0 - 26 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.039 ms | 1 - 30 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 6.646 ms | 1 - 60 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA7255P ADP | SA7255P | TFLITE | 9.047 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA7255P ADP | SA7255P | QNN | 8.951 ms | 1 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.824 ms | 0 - 10 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 1.821 ms | 1 - 3 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8295P ADP | SA8295P | TFLITE | 2.668 ms | 0 - 24 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8295P ADP | SA8295P | QNN | 2.645 ms | 1 - 19 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.832 ms | 0 - 12 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 1.813 ms | 0 - 2 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8775P ADP | SA8775P | TFLITE | 2.756 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8775P ADP | SA8775P | QNN | 2.72 ms | 1 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 4.275 ms | 0 - 32 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 5.742 ms | 1 - 13 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 64.699 ms | 1 - 13 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 9.047 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 8.951 ms | 1 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.818 ms | 0 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 1.798 ms | 1 - 4 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 2.756 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 2.72 ms | 1 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 2.16 ms | 0 - 36 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 2.393 ms | 1 - 31 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.085 ms | 1 - 1 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 11.223 ms | 2 - 2 MB | INT8 | NPU | -- |
## License
* The license for the original implementation of YOLOv11-Detection-Quantized 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: Object Detection](https://docs.ultralytics.com/tasks/detect/)
* [Source Model Implementation](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect)
## 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:[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
|