Yolo-v7-Quantized: Optimized for Mobile Deployment
Quantized real-time object detection optimized for mobile and edge
YoloV7 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 Yolo-v7-Quantized found here.
More details on model performance accross various devices, can be found here.
Model Details
- Model Type: Object detection
- Model Stats:
- Model checkpoint: YoloV7 Tiny
- Input resolution: 720p (720x1280)
- Number of parameters: 6.24M
- Model size: 6.23 MB
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model |
|---|---|---|---|---|---|---|---|---|
| Yolo-v7-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 4.408 ms | 0 - 15 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 4.823 ms | 0 - 11 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 7.453 ms | 0 - 11 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 2.814 ms | 0 - 74 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 3.178 ms | 1 - 55 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 5.413 ms | 2 - 122 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 2.888 ms | 0 - 52 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 3.348 ms | 1 - 49 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 5.178 ms | 0 - 89 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 9.984 ms | 0 - 70 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 13.086 ms | 1 - 8 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 94.686 ms | 4 - 8 MB | INT8 | GPU | -- |
| Yolo-v7-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 4.356 ms | 0 - 2 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 3.739 ms | 1 - 2 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 4.376 ms | 0 - 3 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 3.771 ms | 1 - 3 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | SA8775 (Proxy) | SA8775P Proxy | TFLITE | 4.381 ms | 0 - 144 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | SA8775 (Proxy) | SA8775P Proxy | QNN | 3.78 ms | 1 - 3 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 4.398 ms | 0 - 2 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 3.761 ms | 1 - 2 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | SA8295P ADP | SA8295P | TFLITE | 6.058 ms | 0 - 53 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | SA8295P ADP | SA8295P | QNN | 5.157 ms | 1 - 7 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 5.031 ms | 0 - 78 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 4.638 ms | 1 - 58 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 4.153 ms | 1 - 1 MB | INT8 | NPU | -- |
| Yolo-v7-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 9.17 ms | 7 - 7 MB | INT8 | NPU | -- |
License
- The license for the original implementation of Yolo-v7-Quantized can be found here.
- The license for the compiled assets for on-device deployment can be found here
References
- YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
- Source Model Implementation
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 API (serverless) does not yet support pytorch models for this pipeline type.